JP2008299753A - Advertisement output system, server device, advertisement outputting method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that an advertisement fitting user's preference or the like can not be outputted in an interactive system such as a chat. <P>SOLUTION: In the interactive system such as a chat, an advertisement fitting the user's preference or the like can be outputted by the server device comprising: a server side sentence receiving part for receiving a sentence from an information terminal device; a natural language processing part for performing natural language processing of the sentence received by the sentence receiving part and acquiring a processing result including at least one or more terms; an advertisement information acquisition part for using the processing result acquired by the natural language processing part to acquire advertisement information; and an advertisement information transmitting part for transmitting the advertisement information acquired by the advertisement information acquisition part to the information terminal device. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an advertisement output system that receives text input to a terminal and transmits an advertisement to the terminal in real time.

  In the conventional advertisement output system, with regard to advertisement information of goods or services provided on a computer network represented by the Internet, the hobby preference information for each user is grasped by the reaction and operation for each user's avatar, and individually It aims at providing advertisement information (for example, refer to patent documents 1).

Conventionally, there is a technology that can automatically extract reputation information from the world's blogs and Web diaries. Using this technique, it is possible to analyze the degree of excitement (burst degree) and reputation of keywords (for example, see Non-Patent Document 1).
JP 2006-30482 A (first page, FIG. 1 etc.) Internet <URL: http://www.hottolink.co.jp/service/analyze/bw.html>

  However, in the conventional advertisement output system, preference is obtained from a reaction or operation with respect to the avatar, and thus the preference determination is extremely inaccurate. Further, it has been impossible to effectively advertise that a sentence generated from a user such as a chat system, a TV phone, or a telephone is analyzed and an appropriate advertisement is presented to the user from the analysis result.

  In the technique in Non-Patent Document 1, it is considered that the technique of morpheme analysis is used for Japanese analysis. When the technique of morpheme analysis is used, the colloquial sentence cannot be analyzed with high accuracy. As a result, for example, it is not possible to accurately analyze the reputation of products, services, and objects such as people, such as grammatically correct sentences and chat sentences that frequently appear as inverted sentences. There was a problem. Therefore, it cannot be used when an effective advertisement is presented to a person who inputs a sentence that tends to be in a colloquial tone such as chat.

  The advertisement output system according to the first aspect of the present invention is an advertisement output system comprising an information terminal device and a server device, wherein the information terminal device receives a sentence accepting sentence and a sentence accepted by the sentence accepting part. A terminal-side sentence transmission unit that transmits to the server device; an advertisement information reception unit that receives advertisement information from the server device in response to transmission of the sentence; and an advertisement information output unit that outputs the advertisement information. The server device includes a server-side text receiving unit that receives text from the information terminal device, a natural language processing of the text received by the text receiving unit, and a processing result including at least one term. Using the processing result acquired by the processing unit, the natural language processing unit, an advertisement information acquisition unit that acquires advertisement information, and the advertisement information acquired by the advertisement information acquisition unit as the information terminal device An advertisement output system including the advertisement information transmission unit to be transmitted.

  With such a configuration, it is possible to output appropriate advertisement information for the text input by the user in chat or the like.

  Further, in the advertisement output system of the second invention, the natural language processing unit acquires preference information, which is information related to the preference of the input user of the text, from the text, with respect to the first invention, The advertisement information acquisition unit is an advertisement output system that acquires advertisement information that matches the preference information.

  With this configuration, it is possible to output appropriate advertisement information corresponding to the user's preference information obtained by analyzing the text input by the user in chat or the like.

  Further, in the advertisement output system of the third invention, in contrast to the second invention, the natural language processing unit is a function that is a first character string and information that specifies a function or type of the first character string in a sentence. Word information, a pre-connection symbol that is information specifying the type of character string that can be connected before the first character string, and a type of character string that can be connected after the first character string, or a delimiter A dictionary information storage means for storing one or more dictionary information having a pair of a post-connection symbol that is a symbol indicating the type of the second character string in the sentence delimited by the symbol, and an evaluation target An evaluation information storage means storing at least one evaluation information having an evaluation word, which is a term relating to good or bad, and a first character string for a post-connection symbol indicating that it is a delimiter from the sentences constituting the sentence, Using the first character string as a key, the sentence is divided into two or more second character strings that are two or more character strings, and function word information or a post-connection symbol corresponding to the last first character string of the second character string The sentence division control means for obtaining the information specifying the type of the second character string and acquiring the information specifying the type of the second character string and the second character string, and the sentence division control means In a sentence having a second character string that specifies an evaluation target from the type of the second character string, an evaluation target acquisition unit that acquires the evaluation target from the second character string, and a second character string that includes the evaluation target An evaluation word acquisition unit that specifies a second character string that can include an evaluation word using another second character string type, searches the evaluation information storage unit for an evaluation word included in the second character string, and acquires the evaluation word , The evaluation object acquired by the evaluation object acquisition means, and the evaluation word acquisition hand There using the acquired assessment words, an advertisement output system comprising a processing result configuration means constituting the processing result.

  With this configuration, for example, it is possible to acquire user preference information with high accuracy even in sentences that tend to be colloquial, and to output more appropriate advertisement information corresponding to the user preference information.

  The advertisement output system according to the fourth aspect of the invention relates to the third aspect of the invention, in which the evaluation information storage means compares an evaluation word which is a term relating to the quality of the evaluation target and a score indicating the degree of quality. At least one piece of evaluation information, and the processing result configuration unit is the same as the score acquisition unit that acquires the score paired with the evaluation word from the evaluation information storage unit and the evaluation target acquisition unit. An evaluation calculation means for calculating an evaluation for the evaluation object using one or more scores for the evaluation object, an evaluation object acquired by the evaluation object acquisition means, and a score calculated by the evaluation calculation means An advertisement output system comprising score processing result constituting means for constituting a processing result.

  With this configuration, for example, it is possible to acquire user preference information with high accuracy even in sentences that tend to be colloquial, and to output more appropriate advertisement information corresponding to the user preference information.

  In the advertisement output system according to the fifth aspect of the invention, the natural language processing unit is configured to change a change word, which is a wording word indicating a change in evaluation with respect to the evaluation word, and the evaluation with respect to the fourth invention. Change information storage means for storing one or more pieces of change information that is a pair of evaluation change information that is information of the above, a change word acquisition means for acquiring a change word within a distance of a predetermined number of characters from the evaluation word, An evaluation change information acquisition unit that acquires evaluation change information that is paired with a change word, and a score is changed using the evaluation change information acquired by the evaluation change information acquisition unit with respect to the score acquired by the score acquisition unit. And a change score calculation means for calculating a new score, wherein the evaluation calculation means includes one or more scores for the same evaluation object acquired by the evaluation object acquisition means, and the change score. Calculation means using one or more new score for the same evaluation target calculated is an advertisement output system for calculating an evaluation of the evaluation object.

  With this configuration, in natural language processing, it is possible to perform natural language processing with high accuracy in consideration of subsequent phrases (for example, “not”, etc.), and user preference information can be acquired with high accuracy. More appropriate advertisement information can be output.

  The advertisement output system according to the sixth aspect of the present invention provides the advertisement output system according to any one of the third to fifth aspects, wherein the sentence division control means determines a size (n characters) determined in advance from the head of the sentence constituting the sentence. [N is an integer greater than or equal to 2]), and is a character string that constitutes the third character string that has been cut out. It is determined whether or not there is a character string. If there is no character string of n characters, it is determined whether or not a character string of (n-1) characters from the beginning exists in the dictionary information storage means. This determination is performed by reducing the number of characters one character at a time until the character string exists, or when there is no character string, the determination is performed until the first character is reached and it is determined that the character string exists. If you do, the function word corresponding to the character string When the information, the front connection symbol, and the rear connection symbol are read from the dictionary information storage means, the read function word information, the front connection symbol, and the rear connection symbol are assigned to the character string, and the character string does not exist. Element dividing means for adding unknown word information, which is information indicating an unknown word to a character string, and continuous unknown words to which the element dividing means has added unknown word information. Concatenating characters to form a character string, and consecutive unknown words, concatenating characters that are not hiragana to form a character string, and collecting the concatenated character strings or characters of unknown words that are not concatenated Group information, which is information for specifying the type of character string indicating that the element is divided, reads the adjacent two elements or group information divided by the element dividing unit, and reads the read element or group. The above-mentioned element or group information and the following element or group information can be linked from the preceding element or group information after-connection symbol and the following element or group-information front-connection symbol. Determining whether or not there is an element connecting means for collecting a plurality of second character strings that are the combined character strings by collecting the connectable elements or group information into one character string, and the element connecting means From the function word information corresponding to the last first character string of each of the plurality of second character strings acquired or the post-connection symbol, information specifying the type of each second character string is determined, and each of the plurality of second characters is determined. An advertisement output system comprising character string acquisition means for arranging, on a memory, information for specifying a character string type and a plurality of pairs of information obtained by pairing the plurality of second character strings.

  With this configuration, for example, it is possible to acquire user preference information with high accuracy even in sentences that tend to be colloquial, and to output more appropriate advertisement information corresponding to the user preference information.

  Further, according to the seventh aspect of the present invention, in the advertisement output system according to any one of the first to sixth aspects, the server device includes a sentence storage unit storing sentences and a processing result in the natural language processing unit. A sentence acquisition unit that acquires a sentence according to the sentence from the sentence storage unit, and a server-side sentence transmission unit that transmits the sentence acquired by the sentence acquisition unit to the information terminal device. The advertisement output system further includes a terminal-side text receiving unit that receives the text transmitted by the server device, and a text output unit that outputs the text received by the terminal-side text receiving unit.

  With this configuration, it is possible to output appropriate advertisement information for the text input by the user in the chat performed by the user on the virtual character.

  The advertisement output system according to the eighth aspect of the present invention is an advertisement output system further comprising a second information terminal device as compared with any of the first, fifth to sixth aspects.

  With this configuration, it is possible to output appropriate advertisement information for the text input by the user in the chat performed by the user with another user.

  In the advertisement output system according to the ninth aspect of the present invention, the server device further includes an advertisement information storage unit that stores advertisement information corresponding to the term, according to any one of the first to eighth aspects. The advertisement information acquisition unit is an advertisement output system that acquires, from the advertisement information storage unit, advertisement information corresponding to a term included in a processing result acquired by the natural language processing unit.

  With this configuration, the advertisement information corresponding to the terms included in the processing result acquired by the natural language processing unit can be acquired from the advertisement information storage unit.

  In the advertisement output system according to the tenth aspect of the present invention, the server device further includes an advertisement information storage unit storing advertisement information corresponding to the category, according to any one of the first to eighth aspects. The advertisement information acquisition unit includes a category conversion dictionary storage unit that stores a category conversion dictionary indicating a correspondence between terms and categories, and a category corresponding to a term included in the processing result acquired by the natural language processing unit. An advertisement output system comprising: a category acquisition unit that acquires from the category conversion dictionary storage unit; and an advertisement information acquisition unit that acquires advertisement information corresponding to the category acquired by the category acquisition unit from the advertisement information storage unit. .

  With such a configuration, advertisement information corresponding to a category paired with a term included in the processing result acquired by the natural language processing unit can be acquired from the advertisement information storage unit.

  According to the advertisement output system of the present invention, it is possible to output an advertisement suitable for the user's preference or the like in an interactive system such as chat.

Hereinafter, embodiments of an advertisement output system and the like will be described with reference to the drawings. In addition, since the component which attached | subjected the same code | symbol in embodiment performs the same operation | movement, re-explanation may be abbreviate | omitted.
(Embodiment 1)

  In this embodiment, an advertisement output system that receives text input to a terminal and transmits an advertisement to the terminal in real time will be described. In addition, an advertisement output system that acquires preference information (reputation information), which is information related to user preferences, and outputs advertisement information corresponding to the preference information will be described. In the present embodiment, an advertisement output system in which advertisement information is output to the user's terminal in a system in which the user can enjoy chatting with a virtual character output from the server device will be described. Furthermore, in this Embodiment, the aspect which acquires and outputs advertisement information by matching the target object which user's preference information shows, and the word provided to advertisement information is demonstrated.

  FIG. 1 is a conceptual diagram of an advertisement output system in the present embodiment. The advertisement output system includes an information terminal device 11 and a server device 12. The information terminal device 11 is a terminal used in a system in which a user performs a conversation such as a chat system. The information terminal device 11 is, for example, a personal computer, a mobile terminal, a mobile phone, a landline phone, or the like, which is input by a user when a user interacts with a person or a virtual character, or a voice (a sentence converted into a sound). It is an input device. The server device 12 receives a text input by the user and performs a process for outputting advertisement information that matches the text to the information terminal device 11.

  FIG. 2 is a block diagram of the advertisement output system in the present embodiment. The information terminal device 11 includes a reception unit 1101, an instruction transmission unit 1102, a character reception unit 1103, a character output unit 1104, a text reception unit 1105, a terminal side text transmission unit 1106, a terminal side text reception unit 1107, a text output unit 1108, an advertisement. An information receiving unit 1109 and an advertisement information output unit 1110 are provided.

  The server device 12 includes an advertisement information storage unit 1201, a text storage unit 1202, a character storage unit 1203, an instruction reception unit 1204, a character transmission unit 1205, a server side text reception unit 1206, a natural language processing unit 1207, a text acquisition unit 1208, a server A side sentence transmission unit 1209, an advertisement information acquisition unit 1210, and an advertisement information transmission unit 1211 are provided.

  The accepting unit 1101 accepts various instructions from the user. The instruction is a start instruction for starting an interactive system such as chat. The dialogue system is a system that assists the user of the information terminal device 11 to interact with something. The dialogue system includes a chat with a person, a chat with a virtual character, a telephone call, a TV conference, and the like. The instruction is, for example, a start / end to end the interactive system. The accepting unit 1101 can be realized by a device driver for input means such as a keyboard, control software for a menu screen, or the like.

  The instruction transmission unit 1102 transmits the instruction received by the reception unit 1101 to the server device 12. The instruction transmission unit 1102 is usually realized by a wireless or wired communication unit, but may be realized by a broadcasting unit.

  The character receiving unit 1103 receives a character from the server device 12. The character is, for example, still image or moving image data, moving image and audio data, or the like. The character receiving unit 1103 is usually realized by a wireless or wired communication unit, but may be realized by a broadcast receiving unit.

  The character output unit 1104 outputs the character received by the character receiving unit 1103. Here, the output is usually a display on a display, or a display on the display and an audio output to a speaker. The character output unit 1104 may or may not include an output device such as a display or a speaker. The character output unit 1104 can be implemented by output device driver software, or output device driver software and an output device.

  The text receiving unit 1105 receives text. The sentence receiving unit 1105 receives a sentence at least twice. That is, the information terminal device 11 is used in a situation where the user is interacting. A sentence is a set of one or more sentences. The sentence may be colloquial or incomplete. In addition, the term “acceptance” usually refers to acceptance of texts by manual input, but includes reception from other devices, reading from storage media, and the like. The text input means may be anything such as a keyboard, mouse or microphone. When a sentence is input by voice from the microphone, the information terminal device 11, the server apparatus 12, or another device (not shown) performs a process of converting the voice into a sentence by a voice recognition process. The text receiving unit 1105 can be realized by a device driver for input means such as a keyboard, control software for a menu screen, or the like.

  The terminal-side text transmission unit 1106 transmits the text received by the text reception unit 1105 to the server device 12. The timing which transmits a sentence to the server apparatus 12 does not ask | require. The terminal-side text transmission unit 1106 may transmit the text according to an instruction from the user, or may transmit the text immediately when the text reception unit 1105 receives the text. The terminal-side text transmission unit 1106 is normally realized by a wireless or wired communication unit, but may be realized by a broadcasting unit.

  The terminal-side text receiving unit 1107 receives the text transmitted by the server device 12. The sentence here is a sentence output by the server device 12 and is a sentence transmitted corresponding to the sentence input by the user. The terminal-side text receiving unit 1107 is usually realized by a wireless or wired communication means, but may be realized by a means for receiving a broadcast.

  The text output unit 1108 outputs the text received by the terminal side text receiving unit 1107. Here, the output is usually a display on a display, but includes printing on a printer, sound output, transmission to an external device, and the like. The text output unit 1108 may or may not include an output device such as a display or a speaker. The text output unit 1108 can be implemented by output device driver software, or output device driver software and an output device.

  The advertisement information receiving unit 1109 receives the advertisement information from the server device 12 in response to the transmission of the text by the terminal side text transmitting unit 1106. Note that the advertisement information may exist in the server device 12 or may exist in a third device (not shown). That is, the advertisement information receiving unit 1109 may receive advertisement information from a third device (not shown). In the present embodiment, the advertisement information is mainly described on the assumption that the server device 12 holds the advertisement information. The advertisement information receiving unit 1109 is usually realized by a wireless or wired communication means, but may be realized by a means for receiving a broadcast.

  The advertisement information output unit 1110 outputs the advertisement information received by the advertisement information receiving unit 1109. Here, the output is usually a display on a display, but includes printing on a printer, sound output, transmission to an external device, and the like. The advertisement information output unit 1110 may be considered as including or not including an output device such as a display or a speaker. The advertisement information output unit 1110 may be realized by output device driver software, or output device driver software and an output device.

  The advertisement information storage unit 1201 stores advertisement information corresponding to terms. The advertisement information storage unit 1201 normally stores a plurality of sets of one or more terms and advertisement information. The advertisement information storage unit 1201 may store a plurality of sets of one or more terms and advertisement information pointers, for example. In such a case, the advertisement information may exist in a third device (not shown). The term may be an object for which the user has a positive feeling. The object is information indicating the object. The advertisement information storage unit 1201 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The text storage unit 1202 stores text. Here, the data structure of the stored text does not matter. The sentence here may be a set of terms constituting the sentence. An example of the data structure of the text stored in the text storage unit 1202 will be described later. The text storage unit 1202 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The character storage unit 1203 stores characters. The character is a character that is composed of a still image or a moving image and that virtually interacts with the information terminal device 11 and the user. The character storage unit 1203 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The instruction receiving unit 1204 receives an instruction from the information terminal device 11. The instruction here is, for example, a start instruction. The instruction receiving unit 1204 is usually realized by a wireless or wired communication means, but may be realized by a means for receiving a broadcast.

  The character transmission unit 1205 reads the character from the character storage unit 1203 and transmits it to the information terminal device 11. The character transmission unit 1205 reads the character from the character storage unit 1203 and transmits it to the information terminal device 11 when the instruction receiving unit 1204 receives the start instruction. In addition, the character transmission unit 1205 receives a character from the character storage unit 1203 in accordance with a result of natural language processing on a sentence received by a natural language processing unit 1207, which will be described later, or in accordance with a sentence acquired by the sentence acquisition unit 1208. Read and transmit to the information terminal device 11. The character transmission unit 1205 is usually realized by wireless or wired communication means, but may be realized by broadcasting means.

  The server side text receiving unit 1206 receives text from the information terminal device 11. The server-side text receiving unit 1206 is usually realized by a wireless or wired communication means, but may be realized by a means for receiving a broadcast.

  The natural language processing unit 1207 performs natural language processing on the text received by the server side text receiving unit 1206, and acquires a processing result including at least one or more terms. The processing algorithm in the natural language processing unit 1207 is not limited. For example, the natural language processing unit 1207 may perform morphological analysis on the sentence received by the server-side sentence reception unit 1206 and extract only one or more nouns. In addition, the natural language processing unit 1207, for example, morphologically analyzes the sentence received by the server-side sentence receiving unit 1206, acquires one or more nouns and adjectives that modify the nouns in pairs, and uses the adjectives for the nouns to determine the user The preference for nouns may be determined. In other words, it stores positive (positive) adjectives (for example, “good”, “beautiful”, “beautiful”, “cool”, “great”, etc.) It may be extracted as a term. The natural language processing unit 1207 stores positive adjectives and negative (negative) adjectives (for example, “bad”, “dirty”, “ugly”, “cool”, “no use”), Preference information may be configured by extracting nouns with positive adjectives as positive terms and nouns with negative adjectives as negative terms. The preference information includes object information indicating the object and evaluation information indicating evaluation of the object. The object information is a noun with a positive term or a negative term. The evaluation information is a positive term or a negative term. The natural language processing here may be known natural language processing or natural language processing using functional element analysis described in the second embodiment. The natural language processing unit 1207 can usually be realized by an MPU, a memory, or the like. The processing procedure of the natural language processing unit 1207 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The text acquisition unit 1208 acquires a text corresponding to the processing result in the natural language processing unit 1207 from the text storage unit 1202. The sentence acquisition unit 1208 acquires, for example, a sentence “I like you” from the sentence storage unit 1202 for a sentence “I like cats”. Details of the sentence acquisition algorithm will be described later. The sentence acquisition unit 1208 can be usually realized by an MPU, a memory, or the like. The processing procedure of the sentence acquisition unit 1208 is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The server-side text transmission unit 1209 transmits the text acquired by the text acquisition unit 1208 to the information terminal device 11. The server-side text transmission unit 1209 is usually realized by a wireless or wired communication unit, but may be realized by a broadcasting unit.

  The advertisement information acquisition unit 1210 acquires advertisement information using the processing result acquired by the natural language processing unit 1207. The advertisement information acquisition unit 1210 acquires advertisement information corresponding to the text input by the user of the information terminal device 11. More specifically, for example, the advertisement information acquisition unit 1210 acquires advertisement information that matches the preference information acquired by the natural language processing unit 1207 from the advertisement information storage unit 1201. In addition, the advertisement information acquisition unit 1210 acquires, for example, advertisement information corresponding to the terms included in the processing result acquired by the natural language processing unit from the advertisement information storage unit 1201. Here, the advertisement information that matches the preference information is the advertisement information that the advertisement information acquisition unit 1210 determines to match the preference information. For example, the advertisement information acquisition unit 1210 randomly selects and reads one of the advertisement information managed in a pair with the term (object) acquired by the natural language processing unit 1207. Further, for example, when the preference information acquired by the natural language processing unit 1207 includes an affirmative term, the advertisement information acquisition unit 1210 randomly selects one of the advertisement information that is paired with an object included in the preference information. Select and read. The advertisement information acquisition unit 1210 can usually be realized by an MPU, a memory, or the like. The processing procedure of the advertisement information acquisition unit 1210 is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The advertisement information transmission unit 1211 transmits the advertisement information acquired by the advertisement information acquisition unit 1210 to the information terminal device 11. There is no limitation on the timing at which the advertisement information transmitting unit 1211 receives the advertisement information. However, when the text is received from the information terminal device 11 and the natural language processing or the like is completed, it is preferable to immediately transmit the advertisement information. Further, the advertisement information acquisition unit 1210 acquires an object corresponding to preference information including positive evaluation information (hereinafter also referred to as object information), and the advertisement information acquisition unit 1210 corresponds to the object. When the advertisement information can be acquired, it is preferable to transmit the advertisement information. The advertisement information transmitting unit 1211 is usually realized by a wireless or wired communication unit, but may be realized by a broadcasting unit.

  Next, the operation of the advertisement output system will be described. First, operation | movement of the information terminal device 11 is demonstrated using the flowchart of FIG.

  (Step S301) The reception unit 1101 determines whether or not a start instruction for starting a dialogue system with a character has been received from a user. If a start instruction is accepted, the process proceeds to step S302, and if a start instruction is not accepted, the process returns to step S301.

  (Step S302) The instruction transmission unit 1102 transmits the start instruction received in step S301 to the server device 12. It is assumed that the instruction transmission unit 1102 holds information for communicating with the server device 12. For example, the start instruction includes the URL and IP address of the server device 12. Further, such URL may be information input by the user.

  (Step S303) The character receiving unit 1103 or means (not shown) determines whether or not a character or the like has been received. If a character or the like is received, the process goes to step S304, and if no character or the like is received, the process returns to step S303. The character or the like is, for example, a character and an application screen for performing a chat system (for example, an application screen described in HTML or XML). The means (not shown) is means for receiving an application screen, for example. The character receiving unit 1103 or a means (not shown) is, for example, reception function software included in a so-called Web browser.

  (Step S304) The character output unit 1104 or means (not shown) outputs the character received in step S303. The character or the like is, for example, an application screen for performing a chat system with a character. The character output unit 1104 or means (not shown) is, for example, output function software included in a so-called Web browser. Since the processing until the application screen designated by the user is output is a known technique (for example, a so-called Web browser technique), detailed description thereof is omitted.

  (Step S305) The text receiving unit 1105 determines whether a text has been received from the user. If a sentence is accepted, the process goes to step S306, and if no sentence is accepted, the process goes to step S308.

  (Step S306) The text output unit 1108 outputs the text received in step S305.

  (Step S307) The terminal-side text transmission unit 1106 transmits the text received in step S305. The process returns to step S305.

  (Step S <b> 308) The terminal-side text reception unit 1107 determines whether a text is received from the server device 12. If a sentence is received, the process goes to step S309, and if no sentence is received, the process goes to step S310.

  (Step S309) The text output unit 1108 outputs the text received in step S308. The process returns to step S305.

  (Step S <b> 310) The advertisement information receiving unit 1109 determines whether advertisement information has been received from the server device 12. If the advertisement information is received, the process goes to step S311. If the advertisement information is not received, the process goes to step S312.

  (Step S311) The advertisement information output unit 1110 outputs the advertisement information received in step S310. The process returns to step S305.

  (Step S312) The character receiving unit 1103 determines whether or not a character has been received. If a character is received, the process goes to step S313, and if no character is received, the process goes to step S314.

  (Step S313) The character output unit 1104 outputs the character received in step S312. The process returns to step S305.

  (Step S314) The accepting unit 1101 judges whether or not an instruction to end the chat system has been accepted from the user. If an end instruction is accepted, the process ends. If no end instruction is accepted, the process returns to step S305. It is preferable that means (not shown) transmits an end instruction to the server device 12 before the end of the process.

  Next, the operation of the server device 12 will be described using the flowcharts of FIGS.

  (Step S401) The instruction receiving unit 1204 determines whether a start instruction has been received. If a start instruction is received, the process proceeds to step S402. If a start instruction is not received, the process returns to step S401.

  (Step S402) The character transmission unit 1205 and a means (not shown) read a character and the like. A character or the like is screen information constituting a chat system with a character. It is assumed that the screen information is stored in a storage medium (not shown).

  (Step S403) The character transmission unit 1205 and means (not shown) transmit the character read out in step S402.

  (Step S404) The server-side text receiving unit 1206 determines whether or not a text has been received from the information terminal device 11. If a sentence is received, the process goes to step S405, and if no sentence is received, the process goes to step S404.

  (Step S405) The natural language processing unit 1207 performs natural language processing on the sentence received in step S404, acquires a processing result, and arranges it on the memory. An example of natural language processing will be described with reference to the flowchart of FIG.

  (Step S406) The sentence acquisition unit 1208 acquires a sentence using the processing result obtained in step S405. The sentence acquisition unit 1208 may read the sentence from the sentence storage unit 1202 or may compose the sentence using the processing result. An example of the sentence acquisition process will be described with reference to the flowchart of FIG.

  (Step S407) The server-side text transmission unit 1209 transmits the text acquired in step S406 to the information terminal device 11.

  (Step S408) The advertisement information acquisition unit 1210 determines whether or not to acquire advertisement information using the processing result obtained in step S405. For example, the advertisement information acquisition unit 1210 determines that the advertisement information is acquired when the processing result includes preference information having an affirmative term. Further, for example, the advertisement information acquisition unit 1210 determines that the advertisement information is acquired when two or more identical terms are included in the processing result. In such a case, the advertisement information acquisition unit 1210 recognizes two or more identical terms as important terms. When it is determined that the advertisement information is acquired, the process goes to step S409, and when it is determined that the advertisement information is not acquired, the process goes to step S411.

  (Step S409) The advertisement information acquisition unit 1210 acquires advertisement information using the processing result obtained in step S405, and arranges it on the memory. An example of the advertisement information acquisition process will be described with reference to the flowchart of FIG.

  (Step S410) The advertisement information transmitting unit 1211 transmits the advertisement information acquired in step S409 to the information terminal device 11.

  (Step S411) The character transmission unit 1205 determines whether or not to acquire a character using the processing result obtained in step S405. For example, the character transmission unit 1205 determines that the advertisement information is acquired when the processing result includes preference information having a positive term or a negative term. In addition, the character transmission unit 1205 determines whether the processing result includes an affirmative term (mode 1), an affirmative term is not included, and a negative term is included (mode 2). It is determined whether it is not included (mode 3), and whether it is different from a mode (mode 1, mode 2, or mode 3) corresponding to a processing result that is a result of natural language processing of a previously received sentence. If it is determined that the character is different, the character is acquired. If it is determined that a character is to be acquired, the process proceeds to step S412. If it is determined that no character is to be acquired, the process proceeds to step S404.

  (Step S412) The character transmission unit 1205 acquires the character corresponding to the processing result obtained in step S405 from the character storage unit 1203, and arranges it on the memory. Note that the character corresponding to the processing result is, for example, a character for a mode corresponding to the processing result. In such a case, the character storage unit 1203 holds a mode identifier for identifying a mode and a character as a pair. Moreover, the user can see the character according to the evaluation information (not positive or negative) indicated by the text written by the user by transmitting the character.

  (Step S413) The character transmission unit 1205 transmits the character acquired in step S412 to the information terminal device 11. The process returns to step S404.

  Note that in step S408 in the flowchart of FIG. 4, the advertisement information may always be acquired without determining whether the advertisement information can be acquired.

  Further, in step S411 in the flowchart of FIG. 4, the character may be always acquired without determining whether the character can be acquired.

  Furthermore, in the flowchart of FIG. 4, when the instruction receiving unit 1204 receives an end instruction, the process ends due to power-off or a process end interrupt.

  Next, an example of natural language processing in step S405 will be described using the flowchart in FIG.

  (Step S501) The natural language processing unit 1207 divides the sentence into sentences.

  (Step S502) The natural language processing unit 1207 substitutes 1 for a counter i.

  (Step S503) The natural language processing unit 1207 determines whether or not the i-th sentence exists in the sentence divided in step S501. If the i-th sentence exists, the process proceeds to step S504, and if the i-th sentence does not exist, the process returns to the upper process.

  (Step S504) The natural language processing unit 1207 performs morphological analysis processing on the i-th sentence. As software for performing morphological analysis processing, there is ChaSen (URL: “http://chasen-legacy.sourceforge.jp/”) etc. Since morphological analysis processing is a known technique, description thereof is omitted. By processing, the sentence is divided into terms, and information on the part of speech is obtained.

  (Step S505) The natural language processing unit 1207 acquires a noun term from the analysis result in step S504.

  (Step S506) The natural language processing unit 1207 acquires an affirmative term that is an adjective or an adjective verb from the analysis result in step S504. In other words, the natural language processing unit 1207 includes evaluation information storage means storing one or more positive terms, and the positive information existing in the evaluation information storage means from among the adjectives or adjective verbs in the analysis result. Get terms. In some cases, positive terms cannot be acquired in step S506.

  (Step S507) When the natural language processing unit 1207 acquires an affirmative term in Step S506, the natural language processing unit 1207 configures preference information by pairing the affirmative term and the corresponding noun term, and arranges them on the memory. . The term of the noun corresponding to the positive term is a noun that follows the positive term, or in the case where there is no noun that follows the positive term, for example, the noun that is closest to the positive term. If the positive language cannot be acquired in step S506, the natural language processing unit 1207 places the term acquired in step S505 on the memory. If the positive language cannot be acquired in step S506, the natural language processing unit 1207 may select one term from the two or more terms acquired in step S505 and place it on the memory. These terms may be placed on the memory as they are. There is no limitation on the algorithm for selecting one term from two or more terms. The natural language processing unit 1207 may acquire one term, for example, by a process of acquiring a known important term.

  (Step S508) The natural language processing unit 1207 increments the counter i by one. The process returns to step S503.

  Next, an example of the sentence acquisition process in step S406 will be described using the flowchart of FIG.

  (Step S601) The sentence acquisition unit 1208 determines whether or not preference information (evaluation information or a positive term, but the same meaning) exists in the processing result that is the result of the natural language processing in the natural language processing unit 1207. to decide. If preference information exists, it will go to step S602, and if preference information does not exist, it will go to step S605.

  (Step S602) The sentence acquisition unit 1208 acquires the object (noun) in the preference information and arranges it on the memory.

  (Step S603) The text acquisition unit 1208 acquires a text corresponding to the preference information from the text storage unit 1202. The sentence storage unit 1202 stores, for example, sentences corresponding to preference information and sentences corresponding to terms. The sentence corresponding to the preference information is a sentence in which the next conversation can be continued when the user says “I like <object>”, for example, “Do you like <object>?” Or “I like <object> too!” Or “<object> is good”. The sentence corresponding to the term is "What is <term>?", "What <term>? Of <term>?" The sentence storage unit 1202 holds, for example, a flag “1” indicating that it corresponds to preference information and a sentence, or a flag “0” indicating that it corresponds to a term and a sentence in pairs. Here, when there are a plurality of pieces of preference information in the processing result, the sentence acquisition unit 1208 may randomly select one piece of preference information or may acquire a plurality of pieces of preference information. . In the above, <target> indicates a variable into which the target is inserted.

  (Step S604) The text acquisition unit 1208 inserts the object (noun) acquired in step S602 into the text corresponding to the preference information acquired in step S603. Return to upper process.

  (Step S605) The sentence acquisition unit 1208 selects one term, for example, at random from one or more terms in the processing result. If there is only one term in the processing result, the processing in this step is not necessary. The sentence acquisition unit 1208 may select one important word from a plurality of terms in the processing result. Since a technique for selecting one important word is a known technique, a description thereof will be omitted. A technique for selecting one important word from a plurality of terms is, for example, software for extracting terminology “TermExtract” (see URL “http://gensen.dl.itc.u-tokyo.ac.jp/”), etc. There is. In addition, the sentence acquisition unit 1208 may extract a topic (term) using a known technique for identifying a topic from a sentence.

  (Step S606) The sentence acquisition unit 1208 acquires a sentence corresponding to the term.

  (Step S607) The text acquisition unit 1208 inserts the term acquired in Step S605 into the text acquired in Step S606. Return to upper process.

  In the flowchart of FIG. 6, the sentence acquisition process may use only the target object without using the preference information, for example. In such a case, the sentence acquisition unit 1208 acquires the term of the object and reads the sentence corresponding to the term.

  Next, an example of the advertisement information acquisition process in step S409 will be described using the flowchart in FIG.

  (Step S <b> 701) The advertisement information acquisition unit 1210 determines whether or not preference information exists in the processing result that is the result of the natural language processing in the natural language processing unit 1207. If preference information exists, it will go to step S702, and if preference information does not exist, it will go to step S704.

  (Step S702) The advertisement information acquisition unit 1210 acquires the object (noun) in the preference information and places it on the memory.

  (Step S703) The advertisement information acquisition unit 1210 acquires the advertisement information corresponding to the preference information from the advertisement information storage unit 1201. In such a case, the advertisement information storage unit 1201 stores one or more pieces of information having a target and advertisement information as a pair. This advertisement information is information about an advertisement that the user who is positive about the object wants to present. Return to upper process.

  (Step S704) The advertisement information acquisition unit 1210 randomly selects one term from one or more terms in the processing result, for example. If there is only one term in the processing result, the processing in this step is not necessary. Moreover, the advertisement information acquisition unit 1210 may select one important word from a plurality of terms in the processing result.

  (Step S705) The advertisement information acquisition unit 1210 acquires advertisement information corresponding to the term selected in step S704, and arranges it on the memory. Return to upper process.

  In the flowchart of FIG. 7, the advertisement information acquisition process may use only the target object without using the preference information, for example. In such a case, the advertisement information acquisition unit 1210 acquires the term of the target object and reads a sentence corresponding to the term.

  Moreover, in the flowchart of FIG. 7, when using preference information, the preference information positive with respect to the target object was used. However, negative preference information for the object may be used. That is, the advertisement information storage unit 1201 may store advertisement information that is output to the user when a certain object is negative.

  Hereinafter, a specific operation of the advertisement output system in the present embodiment will be described. FIG. 1 is a conceptual diagram of the advertisement output system.

  Now, the advertisement information storage unit 1201 of the server device 12 stores an advertisement information management table as shown in FIG. The advertisement information management table manages a plurality of records having attribute values of “ID”, “positive object”, “term”, and “advertisement information”. “ID” is an attribute in which a value for identifying a record is stored, and exists for table management. The “positive object” is an attribute that can store one or more positive objects. “Term” is an attribute that can store one or more terms. “Advertisement information” is an attribute that can store one or more file names storing advertisement information. The “advertisement information” may store the advertisement information itself. The advertisement information storage unit 1201 also stores entity data (file) corresponding to the file name indicated by “advertisement information”. The advertisement information may have any data structure such as a character string, a still image, a moving image, a sound, and a combination of a moving image and sound. In the record of “ID = 1” in the advertisement information management table of FIG. 8, if it is determined that the user is positive for “car”, “car”, or “Company A (automobile manufacturer)”, the advertisement information “ A still image of “Company A-car.gif” is output to the information terminal device 11 of the user. In addition, the record of “ID = 2” in the advertisement information management table of FIG. 8 outputs a video of “Company B-Travel Plan.WMV” to the information terminal device 11 of the user who had a conversation in which the term “holiday” appeared. Indicates to do.

  The text storage unit 1202 stores a text management table as shown in FIG. The sentence management table holds a plurality of records having attribute values of “ID”, “type”, and “sentence”. Here, the “type” can take two types of values: “preference information (eg,“ 1 ”)” or “term” (eg, “0”). The “sentence” can store one or more sentences. A sentence may also contain variables. The variable is, for example, <object> or <term>, and “object” or “term” included in the processing result obtained by analysis by the natural language processing unit 1207 can be substituted. The record of “ID = 1” in the sentence management table of FIG. 9 includes “preference information (information about a pair of an object and an affirmative term)” in the processing result obtained by analysis by the natural language processing unit 1207. The sentence “Why do you like <object>?” Is acquired. Further, the record of “ID = 4” in the sentence management table of FIG. 9 does not know the sentence “<term>” when “term” is included in the processing result obtained by the analysis by the natural language processing unit 1207. Yo ~ .Tell me! "

  The character storage unit 1203 stores the character management table shown in FIG. The character management table holds a plurality of records having attribute values of “ID”, “mode”, and “character”. The “mode” can take any of the values “1”, “2”, and “3”. The mode “1” is positive, the mode “2” is negative, and the mode “3”. Indicates that it is neither positive nor negative. In this mode, whether the preference information included in the processing result obtained by the analysis by the natural language processing unit 1207 includes a positive term (mode 1) or a negative term (mode 2). It is determined depending on whether it is not included (mode 3).

  Further, it is assumed that the natural language processing unit 1207 holds an evaluation information management table having positive terms such as “good”, “like”, “love”, “great”, “beautiful”.

  In this situation, it is assumed that the user of the information terminal device 11 inputs an instruction to start the “advertisement chat system” to the information terminal device 11 in order to enjoy chatting with the virtual character of the server device 12. The start instruction includes, for example, a URL indicating a portal screen of the “advertisement chat system” of the server device 12. The “advertising chat system” is a system that can interact with a character.

  Next, the reception unit 1101 of the information terminal device 11 receives a start instruction to start the advertising chat system from the user. Then, the instruction transmission unit 1102 transmits the received start instruction to the server device 12 according to the URL included in the start instruction.

  Next, the instruction receiving unit 1204 of the server device 12 receives the start instruction. Then, a means (not shown) of the server device 12 reads a file (for example, described in HTML) defining a portal screen of the advertisement chat system from a storage medium (not shown) and transmits the file to the information terminal device 11. Also, the character transmission unit 1205 reads the mode 3 character “normal dog.WMV” from the character management table of FIG. 10 and transmits it to the information terminal device 11. It is assumed that the character “normal dog.WMV” in mode 3 is selected as the initial character of chat.

  Next, means (not shown) of the information terminal device 11 receives the data of the portal screen of the advertisement chat system from the server device 12, interprets and executes it, and displays it on the display. In addition, the character receiving unit 1103 receives the character “normal dog.WMV” from the server device 12. Then, the character output unit 1104 outputs the character “normal dog.WMV”. FIG. 11 shows an example of output of the portal screen and the character “normal dog.WMV”. In FIG. 11, the name of the dog character is “X-kun”. On the portal screen, “X-kun” and the user of the information terminal device 11 interact with each other. The screen also has an area (right side) for outputting advertisements. Also, in FIG. 11, “X-kun” remarks saying “Now, let's start talking” is displayed. Such remarks are information that is defined in the portal screen and transmitted by being included in the portal screen.

  Then, the user of the information terminal device 11 thinks that he / she will enjoy chatting with “X-kun” (let's bully “X-kun” a little), and “I like cats,” says “ Suppose you entered it in the "Your remarks" field. Then, the text receiving unit 1105 receives the text “I like cats, but I do”. Next, the sentence output unit 1108 outputs the accepted sentence to the “Your Remarks” field. Then, the terminal-side text transmission unit 1106 transmits the accepted text “I like cats, but I do”.

  Next, the server-side text receiving unit 1206 receives the text “I like cats, but I like” from the information terminal device 11. Next, the natural language processing unit 1207 tries to divide the sentence into sentences. However, since the sentence “I like cats” is a single sentence, the sentence is placed in the memory as it is.

  Next, the natural language processing unit 1207 performs a morphological analysis process on the sentence “I like cats,” and divides the sentence into morphemes, such as “I: noun”, “neko: noun”, “like: adjective verb”, etc. Get the information. Next, the natural language processing unit 1207 acquires the noun terminology and the noun terms “me” and “cat”. Next, the natural language processing unit 1207 acquires the affirmative term “like” which is an adjective verb by using the affirmative term information in the evaluation information management table. In other words, the natural language processing unit 1207 first acquires the adjective verb “like” and checks whether the “like” exists in the evaluation information management table. Since “like” exists in the evaluation information management table, the natural language processing unit 1207 configures preference information “cat, like” and arranges it on the memory. The natural language processing unit 1207 configures preference information “cat, likes” in association with the noun “cat” close to the adjective verb “like”. The natural language processing unit 1207 places the remaining noun “I” on the memory as a term. Here, the preference information “cat, likes” and the term “I” are the processing results of the natural language processing unit 1207. The processing result may include a flag (for example, “1”) indicating preference information, and the processing result may be “1, cat, like”. Note that the natural language processing unit 1207 may be configured such that a term in the processing result includes a flag such as “0, I”. Here, it is assumed that the natural language processing unit 1207 has obtained the processing result “1, cat, likes”, “0, me”.

  Next, the sentence acquisition unit 1208 determines whether or not preference information exists in the processing result that is the result of the natural language processing in the natural language processing unit 1207. Here, the text acquisition unit 1208 detects “1” included in the processing result, and determines that the preference information exists in the processing result. Then, the sentence acquisition unit 1208 acquires the object “cat” in the preference information and places it on the memory. Next, the sentence acquisition unit 1208 acquires the sentence “Why do you like <object>” from the sentence storage unit 1202 corresponding to the preference information. Here, the sentence acquisition unit 1208 selects a sentence that matches the type (whether it is preference information or a term) from the sentence management table of FIG. 9, and when two or more sentences exist, sequentially selects one sentence, Suppose you want to get. Next, the sentence acquisition unit 1208 inserts the acquired object “cat” into the acquired sentence “Why do you like <object>”, and obtains “Why do you like cats?”. .

  Next, the server-side text transmission unit 1209 transmits the acquired text “Why do you like cats?” To the information terminal device 11.

  Next, the terminal-side text receiving unit 1107 receives the text “Why do you like cats?” From the server device 12. Then, the text output unit 1108 receives the received text “Why do you like cats?”. It is output to the field of “X-kun's remark” (see 1201 in FIG. 12).

  Next, the advertisement information acquisition unit 1210 determines whether or not to acquire the advertisement information by using the obtained processing result “1, cat, likes” “0, I”. Here, it is assumed that the advertisement information acquisition unit 1210 determines that the advertisement information is acquired when the preference information is included. Then, the advertisement information acquisition unit 1210 determines that the advertisement information is acquired from “1” of “1, cat, like”. Next, the advertisement information acquisition unit 1210 searches the advertisement information management table using the object “cat” of the processing result “1, cat, likes” as a key, and reads the advertisement information “cat goods.gif”. Then, the advertisement information transmitting unit 1211 transmits the advertisement information “cat goods.gif” to the information terminal device 11.

  Next, the advertisement information receiving unit 1109 receives the advertisement information “cat goods.gif” from the server device 12. Then, the advertisement information output unit 1110 outputs the advertisement information “cat goods.gif” to be added to a predetermined advertisement area (see 1202 in FIG. 12). Since the advertisement information “cat goods.gif” is the first advertisement information, it is output at the head of the advertisement area in FIG.

  Next, the character transmission unit 1205 determines whether or not to acquire a character by using the obtained processing results “1, cat, likes”, “0, me”. That is, the character transmission unit 1205 determines that “1, cat, like” includes an affirmative term (mode 1) and changes from the current mode, mode 3, so that the character is changed. It is determined that it will be acquired. Then, the character transmission unit 1205 acquires the character “laughing dog.WMV” in mode 1 corresponding to the processing result “1, cat, like” from the character management table. Then, the character transmission unit 1205 transmits the acquired character “laughing dog.WMV” to the information terminal device 11. It is assumed that the server device 12 holds the current mode on the memory.

  Next, the character receiving unit 1103 receives the character “laughing dog.WMV”. Then, the character output unit 1104 outputs the received character “laughing dog.WMV” (see 1203 in FIG. 12).

  After that, the user enjoys chatting with “X-kun” (inputs a sentence and reads X-kun's remarks), and finally says “X company is good”. Then, a natural language process similar to the process described above is performed, and an advertisement of “Company X—Product list.gif” shown at 1204 in FIG. 12 is displayed on the screen of the information terminal device 11.

  Then, it is assumed that the user has pressed the “end” button in FIG. Then, the chat system ends.

  As described above, according to the present embodiment, the user can obtain advertisement information that matches his / her preference while interacting with the character. In the situation where the user is chatting, the user always stares at the screen. Therefore, if the advertisement information is output while staring at the screen, the advertising effect is greatly improved. In addition, since the text input by the user in the dialogue is analyzed in a natural language and the user's preference is acquired, according to the present embodiment, the user's exact preference is obtained and an advertisement is displayed during the dialogue. Therefore, the user's attention is very easy to call out.

  In addition, according to this Embodiment, although the chat system was a chat system with a virtual character, a chat system with another user etc. may be sufficient. The same applies to other embodiments.

  Further, according to the present embodiment, the advertisement information is held by the server device. However, the advertisement information may be held by another third device, and the advertisement information may be transmitted to the information terminal device 11 from the third device via the server device or directly from the third device. In other words, the following advertisement output method may be realized by this system. The method is an advertisement output method realized by a sentence receiving unit, a natural language processing unit, an advertisement information acquiring unit, and an advertisement information transmitting unit, and receives a sentence from an information terminal device, The natural language processing is performed on the text received in the text receiving step, the processing result including at least one term is acquired, and the processing result acquired in the natural language processing step is used to acquire the advertisement information. An advertisement output method comprising: an advertisement information acquisition step; and an advertisement information transmission step of transmitting the advertisement information acquired in the advertisement information acquisition step to the information terminal device.

  Further, in the natural language processing step of the above method, preference information that is information related to the preference of the input user of the text is acquired from the text, and advertisement information that matches the preference information is acquired in the advertisement information acquisition step. That is preferred.

  Furthermore, the processing in the present embodiment may be realized by software. Then, this software may be distributed by software download or the like. Further, this software may be recorded on a recording medium such as a CD-ROM and distributed. This also applies to other embodiments in this specification. Note that the software that implements the server device in the present embodiment is the following program. In other words, this program performs natural language processing on a computer-side server-side text receiving unit that receives text from an information terminal device and text received by the server-side text receiving unit, and generates a processing result including at least one term. Using the acquired natural language processing unit, the processing result acquired by the natural language processing unit, the advertising information acquisition unit for acquiring advertising information, and the advertising information acquired by the advertising information acquisition unit are transmitted to the information terminal device It is a program for functioning as an advertisement information transmitting unit.

Further, in the above program, the natural language processing unit acquires preference information that is information related to the preference of the input user of the text from the text, and the advertisement information acquisition unit acquires advertisement information that matches the preference information. It is preferred to have the computer function to obtain.
(Embodiment 2)

  In the present embodiment, an advertisement output system that transmits an advertisement to a terminal in an interactive system such as a chat, using morphological analysis, using a natural language analysis technique called functional element analysis, obtaining preference information, An advertisement output system that matches the preference information will be described. In functional element analysis, a function word stored in advance is used to divide a sentence into elements and output them. In functional element analysis, a sentence is divided into, for example, three types of terms: a subject, a subtitle, and an operation state. In addition, when obtaining preference information, a technology that accurately obtains preference information by using a wording (such as “No”) that follows an evaluation word (such as “Good”, “Like”, “Hate”) that indicates preference Is also explained. FIG. 1 is a conceptual diagram of an advertisement output system in the present embodiment.

  The advertisement output system includes an information terminal device 11 and a server device 132. FIG. 13 is a block diagram of the server device 132 constituting the advertisement output system in the present embodiment.

  The server device 132 includes an advertisement information storage unit 13201, a text storage unit 1202, a character storage unit 1203, an instruction reception unit 1204, a character transmission unit 1205, a server side text reception unit 1206, a natural language processing unit 13207, a text acquisition unit 1208, a server A side sentence transmission unit 1209, an advertisement information acquisition unit 13210, and an advertisement information transmission unit 1211 are provided.

  The natural language processing unit 13207 includes a dictionary information storage unit 132070, an evaluation information storage unit 132071, a change information storage unit 132072, a sentence division control unit 132073, an evaluation target acquisition unit 132074, an evaluation word acquisition unit 132075, a processing result configuration unit 132076, a change Word acquisition means 132077, evaluation change information acquisition means 132078, and change score calculation means 13209.

  The sentence division control unit 132073 includes an element division unit 1320731, an element connection unit 1320732, and a character string acquisition unit 1320733.

  The processing result configuration unit 132076 includes a score acquisition unit 1320761, an evaluation calculation unit 1320762, and a score processing result configuration unit 1320763.

  The advertisement information storage unit 13201 stores advertisement information corresponding to the evaluation for the object. The advertisement information storage unit 13201 has a pair of an object, a condition including a score indicating evaluation (for example, “+3 or more”, etc.), and advertisement information. The advertisement information storage unit 13201 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The natural language processing unit 13207 performs natural language processing on the text received by the server side text receiving unit 1206, and acquires a processing result including at least one term. Here, in particular, the natural language processing unit 13207 acquires preference information, which is information related to the preference of the input user of the text, from the text by using a natural language analysis technique using a technique called functional element analysis. The natural language processing unit 13207 may acquire a user's evaluation of the object as a score using a natural language analysis technique using a functional element analysis technique. A specific example of the processing of the natural language processing unit 13207 will be described later. The technique of functional element analysis is a technique for dividing a sentence, and it goes without saying that the technique of functional element analysis can be used even for sentences for which preference information cannot be obtained.

  The dictionary information storage means 132070 stores one or more dictionary information. Here, it can be said that the dictionary information is information for separating sentences so that evaluation can be easily obtained from the sentence to be analyzed. The dictionary information includes a first character string, function word information, a front connection symbol, and a back connection symbol. The first character string consists of one or more characters. The first character string includes a function word indicating a function in an agglutinative sentence such as “to”, “ni”, and “da”. The function word information is information that identifies the function or type of the first character string in the sentence. The pre-connection symbol is information that identifies the type of character string that can be connected before the first character string. The post-connection symbol is a type of character string that can be connected after the first character string, or a symbol indicating that it is a delimiter, and is information that identifies the type of the second character string in the sentence delimited by the symbol. is there. The second character string is a character string in which the first character string or two or more first character strings are combined. The type of the second character string is, for example, “sentence of sentence”, “subtitle of sentence”, “operation state”, “quasi-theme”, and the like. The “sentence subject” indicates a character string indicating the subject of the sentence to be analyzed. “Sentence subtitle” indicates a character string indicating the subtitle of the sentence to be analyzed. “Operation state” indicates a character string indicating some operation or state. The “operation state” may be classified into two, such as “operation” and “state”. “Sub-theme” indicates a sub-theme. A specific example of dictionary information will be described later. The dictionary information storage unit 132070 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The evaluation information storage unit 132071 stores one or more pieces of evaluation information having evaluation words that are terms relating to the quality of evaluation targets. The evaluation information may be only the evaluation word, may be information of a set of flags indicating whether the evaluation word is positive or negative, or may be information having a pair of scores indicating the evaluation word and the degree of good or bad. The evaluation words are, for example, “like”, “dislike”, “good”, “bad”, “no”, “nice”, “great”, and the like. The score is usually a numerical value such as “−5 to +5”, but may be a rank such as “A, B, C” or the like. The evaluation information storage unit 132071 is preferably a nonvolatile recording medium, but can also be realized by a volatile recording medium.

  The change information storage unit 132072 stores one or more pieces of change information that is information for changing the evaluation. Further, the change information storage means 132072 stores one or more change information that is a pair of a change word that is a wording indicating an evaluation change with respect to the evaluation word and evaluation change information that is information for changing the evaluation. Yes. The change information is a pair of a change word and evaluation change information. The change word is a wording that indicates a change in evaluation with respect to the evaluation word. The change word is, for example, “not”, “not”, “cannot say” or “too much”. That is, if a positive evaluation word “like” is followed by a change word “not”, the evaluation must be negative, not positive. The evaluation change information is information for changing the evaluation. Specifically, it can be said that the evaluation change information is information for changing the score. The evaluation change information is, for example, a numerical value such as “−1” or “0.5”. That is, for example, when the score “+4” is set for the evaluation word “like”, the evaluation change information for the change word “not” is “−1”, and the evaluation change information for the change word “too” is “0.5”. " Then, the modified score calculation unit 13209 described later multiplies each numerical value to calculate “4 × (−1) × (0.5) = − 2”. Here, “+5” is very positive, “−5” is very negative, and “I do not like” is “−2”, which is slightly negative. As change words, words that enhance evaluation (for example, “very” or “very”), words that weaken evaluation (for example, “too much”), words that reverse evaluation (for example, “not”), and invalidation There are different types of words (for example, “it cannot be said”). For example, a word that enhances evaluation is given evaluation change information larger than “1” depending on the degree of strengthening, and a word that weakens evaluation is given, for example, evaluation change information smaller than “1” depending on the degree of weakening. The word that reverses the evaluation is given "-1". The change information storage unit 132072 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The sentence division control unit 132073 acquires information specifying the second character string and the type of the second character string from the sentences constituting the sentence. More specifically, the sentence division control unit 132073 takes out a first character string corresponding to a post-connection symbol indicating a delimiter from sentences constituting the sentence, and uses the first character string as a key to make the sentence two or more. Divide into two or more second character strings that are character strings, and obtain information specifying the type of the second character string from the function word information or the post-connection symbol corresponding to the last first character string of the second character string Then, information specifying the second character string and the type of the second character string is acquired.

  The element division unit 1320731 constituting the sentence division control unit 132073 cuts out a third character string having a predetermined size (n characters [n is an integer of 2 or more]) from the head of the sentence constituting the sentence, and cuts out the third character string. A character string that constitutes a three-character string, in which the character string of n characters from the beginning is judged in order from the first character string to determine whether or not the character string exists in the dictionary information storage unit 132070. Determines whether or not a character string of (n-1) characters from the beginning exists in the dictionary information storage means 132070, and performs such determination until the character string exists by reducing the number of characters one character at a time, Alternatively, if the character string does not exist, the determination is performed until the first character is reached. If it is determined that the character string exists, the function word information corresponding to the character string, the pre-connection symbol, and the post-connection are provided. symbol Is read from the dictionary information storage means 132070, and the read function word information, the front connection symbol, and the back connection symbol are assigned to the character string, and if there is no character string, the character string of one character is unknown. Unknown word information, which is information indicating a word, is added. The element dividing unit 1320731 can be usually realized by an MPU, a memory, or the like. The processing procedure of the element dividing unit 1320731 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The element connecting unit 1320732 is a continuous unknown word to which the element dividing unit 1320731 has added unknown word information, connects hiragana characters to form a character string, is a continuous unknown word, and connects characters that are not hiragana The character string is composed, and the concatenated information is specified to identify the type of character string indicating the concatenation for the concatenated character string or the character of the unknown word that is not concatenated. The element dividing unit 1320731 reads out the divided element or group information, and among the read out element or group information, the preceding element or group information has a subsequent connection symbol, and the subsequent element or group information has a previous connection symbol. Judgment is made on whether or not the preceding element or group information can be linked to the later element or group information. Or collectively grouped information into a single string, a plurality of second string is the combined string, obtains. The element coupling unit 1320732 can be realized typically as an MPU, a memory, or the like. The processing procedure of the element connecting means is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The character string acquisition unit 1320733 identifies the type of each second character string from the function word information or the post-connection symbol corresponding to the last first character string of each of the plurality of second character strings acquired by the element connection unit 1320732. Information is determined, information for specifying the types of the plurality of second character strings, and a plurality of pairs of information in which the plurality of second character strings are paired are arranged on the memory. The character string acquisition unit 1320733 is usually realized by an MPU, a memory, or the like. The processing procedure of the character string acquisition unit 1320733 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The evaluation target acquisition unit 132074 specifies the second character string including the evaluation target from the type of the second character string acquired by the sentence division control unit 132073, and acquires the evaluation target from the second character string. For example, the sentence “Company A is a very good company” is divided by the sentence division control unit 132073 into “theme“ Company A is ”and the operation state“ is a very good company ””. And the evaluation object acquisition means 132074 specifies that the theme “Company A is” is the second character string including the evaluation object. Next, the evaluation target acquisition unit 132074 acquires the term “Company A” included in the subject “Company A” and sets it as an evaluation target. The evaluation target acquisition unit 132074 can be usually realized by an MPU, a memory, or the like. The processing procedure of the evaluation target acquisition unit 132074 is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The evaluation word acquisition unit 132075 specifies a second character string that can include an evaluation word by using another second character string type in the sentence having the second character string including the evaluation target, and the second character string has An evaluation word is retrieved from the evaluation information storage means 132071 and acquired. The evaluation word acquisition unit 132075 can usually be realized by an MPU, a memory, or the like. The processing procedure of the evaluation word acquisition unit 132075 is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The processing result configuration unit 132076 configures the processing result using the evaluation target (same as the object) acquired by the evaluation target acquisition unit 132074 and the evaluation word acquired by the evaluation word acquisition unit 132075. The processing result may be only the evaluation target, or may be information having the evaluation target and the score (positive degree, negative degree). The data structure of the processing result does not matter. The processing result configuration unit 132076 can be typically realized by an MPU, a memory, or the like. The processing procedure of the processing result configuration means is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The score acquisition unit 1320761 constituting the processing result configuration unit 132076 acquires a score paired with the evaluation word from the evaluation information storage unit 132071. The score acquisition unit 1320761 is usually realized by an MPU, a memory, or the like. The processing procedure of the score acquisition means 1320761 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The evaluation calculation unit 1320762 calculates an evaluation for the evaluation target using one or more scores for the same evaluation target acquired by the evaluation target acquisition unit 132074. Further, with respect to the evaluation object, one or more scores for the same evaluation object acquired by the evaluation object acquisition unit 132074 and one or more new scores for the same evaluation object calculated by the change score calculation unit 13209 are used. The evaluation of is calculated. The evaluation calculation unit 1320762 can usually be realized by an MPU, a memory, or the like. The processing procedure of the evaluation calculation means 1320762 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The score processing result configuring unit 1320763 configures a processing result using the evaluation object acquired by the evaluation object acquiring unit 132074 and the score calculated by the evaluation calculating unit 1320762. The score processing result configuring unit 1320663 can be realized typically as an MPU, a memory, or the like. The processing procedure of the score processing result constituting unit 1320663 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The change word acquisition unit 132077 acquires the change word within the distance of the predetermined number of characters from the evaluation word using the change word information included in the change information of the change information storage unit 132072. Needless to say, the change word may not exist. The change word acquisition means 132077 can be usually realized by an MPU, a memory, or the like. The processing procedure of the changed word acquisition unit 132077 is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The evaluation change information acquisition unit 132078 acquires the evaluation change information paired with the change word acquired by the change word acquisition unit 132077 from the change information storage unit 132072. The evaluation change information acquisition unit 132078 can usually be realized by an MPU, a memory, or the like. The processing procedure of the evaluation change information acquisition unit 132078 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The change score calculation unit 13209 changes the score using the evaluation change information acquired by the evaluation change information acquisition unit 132078 with respect to the score acquired by the score acquisition unit 1320761, and calculates a new score. The arithmetic expression of the change score calculation means 13209 does not matter. A specific example of the calculation method will be described later. The change score calculation means 13209 can be usually realized by an MPU, a memory, or the like. The processing procedure of the modified score calculation means 13209 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The advertisement information acquisition unit 13210 acquires advertisement information using the processing result acquired by the natural language processing unit 1207. The advertisement information acquisition unit 13210 searches the advertisement information storage unit 1201 using the target object and score, which are processing results, as keys, and acquires advertisement information that matches the conditions using the target object and score as parameters. Note that the advertisement information storage unit 1201 stores advertisement information in pairs with conditions (for example, “Company A, rank> 2.5”, etc.) using the object and the score as parameters. The advertisement information acquisition unit 13210 can be usually realized by an MPU, a memory, or the like. The processing procedure of the advertisement information acquisition unit 13210 is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  Next, the operation of the advertisement output system will be described. This advertisement output system differs from the advertisement output system of Embodiment 1 in natural language processing and advertisement acquisition processing. Therefore, among the operations of the server device 132 constituting the advertisement output system, the natural language processing of the present embodiment corresponding to step 405 of the first embodiment will be described using the flowcharts of FIGS. 14 to 19. This natural language processing uses a natural language analysis technique called functional element analysis. Note that description of steps already described in the flowchart of FIG. 14 and the like is omitted.

  (Step S1401) The sentence division control unit 132073 performs sentence division control processing for the i-th sentence. The sentence division control process will be described in detail with reference to the flowchart of FIG.

  (Step S1402) The evaluation target acquisition unit 132074 specifies the second character string including the evaluation target from the type of the second character string acquired by the sentence division control unit 132073, and acquires the evaluation target from the second character string. .

  (Step S1403) The evaluation word acquisition unit 132075 specifies a second character string that can include an evaluation word by using another second character string type in the sentence having the second character string including the evaluation target, and the second character string. The evaluation word included in the character string is retrieved from the evaluation information storage unit 132071 and acquired. The evaluation word acquisition unit 132075 acquires a second character string corresponding to another specific second character string type (for example, “state”) in the sentence having the second character string including the evaluation target. The “state” may be an “operating state”.

  (Step S1404) The processing result constituting unit 132076 reads one or more evaluation words of the evaluation information storage unit 132071 in order to determine whether or not the evaluation word is included in the second character string acquired in Step S1403. The second character string is searched using each evaluation word as a key.

  (Step S1405) The processing result constituting unit 132076 determines whether or not the second character string includes an evaluation word as a result of the search in step S1404. If the evaluation word is included, the process goes to step S1406. If the evaluation word is not included, the process goes to step S1410.

  (Step S1406) The score acquisition unit 1320761 reads the score corresponding to the evaluation word from the evaluation information storage unit 132071.

  (Step S1407) The change score calculation means 13209 performs a score change process. Details of the score change processing will be described with reference to the flowchart of FIG.

  (Step S1408) The modified score calculation means 13209 temporarily stores the calculated score.

  (Step S1409) The evaluation calculation means 1320762 uses the one or more scores for the same evaluation object acquired by the evaluation object acquisition means 132074 in the past from the same user's remarks, and evaluates the evaluation object (score ) Is calculated. Such processing is called cumulative score calculation processing. That is, the evaluation calculation unit 1320762 calculates the evaluation score of the user's object in consideration of the past remarks of the user. The cumulative score calculation process is, for example, a process of calculating an average value of scores so far. Further, the cumulative score calculation process is a process for calculating a weighted average value of scores so far, for example. It is preferable that the weighted average is calculated so that the closer the score in time, the greater the weight.

  (Step S1410) The evaluation output device 13 increments the counter i by 1. The process returns to step S503.

  (Step S1411) The score processing result configuring unit 1320766 configures the processing result using the evaluation object acquired by the evaluation object acquiring unit 132074 and the score calculated by the evaluation calculating unit 1320762. The data structure of the processing result does not matter. An example of the data structure of the processing result will be described later. Return to upper process.

  In the flowchart of FIG. 14, “evaluation target (also referred to as subject or object)”, “state (also referred to as operation state)”, etc. acquired in the middle of the process are additionally written in the memory and retained. , And.

  Next, the sentence division control process in step S1401 will be described in detail with reference to the flowchart in FIG.

  (Step S1501) The element dividing means 1320731 substitutes 1 for the counter i.

  (Step S1502) The element dividing unit 1320731 determines whether or not the i-th character exists in the sentence to be analyzed. If the i-th character exists, the process goes to step S1503. If the i-th character does not exist, the process goes to step S1513.

  (Step S1503) The element dividing unit 1320731 acquires a character string from the i-th character to the (i + n) -th character in the sentence to be analyzed, and puts it on the buffer. n is a fixed number, for example, “5”. Note that if the i th character to the (i + n) th character do not exist in the sentence to be processed, the element dividing unit 1320731 acquires the i th character to the last character of the sentence and puts them on the buffer. That is, the element dividing unit 1320731 obtains a character string from the i-th character to the (i + n) -th character and puts it on the buffer.

  (Step S1504) The element division unit 1320731 searches the dictionary information storage unit 132070 for the character string placed on the buffer in step S1503.

  (Step S1505) The element division unit 1320731 determines whether or not the character string arranged on the buffer in step S1503 is present in the dictionary information storage unit 132070 in the search result in step S1504. If the character string exists, the process goes to step S1510. If the character string does not exist, the process goes to step S1506. The case where a character string exists is a case where a character string that matches the character string arranged on the buffer exists in the first character string included in the dictionary information.

  (Step S1506) The element dividing unit 1320731 determines whether or not the length of the character string placed on the buffer in step S1503 is one character. If it is one character, go to step S1508, and if it is not one character, go to step S1507.

  (Step S1507) The element dividing unit 1320731 deletes the last character from the character string arranged on the buffer in Step S1503, reduces the character by one, obtains the character string from the i-th, and arranges it on the buffer. The element dividing unit 1320731 may simply delete the last character on the buffer. The process returns to step S1504.

  (Step S1508) The element dividing unit 1320731 assigns an “unknown word” flag, which is information indicating an unknown word, to the i-th character. “Assigning an“ unknown word ”flag” means that the i-th character can be distinguished from other characters or character strings. For example, the i-th character that is an unknown word is given some flag to other characters or character strings. It is not necessary to give anything to the characters. Furthermore, there is a buffer for storing an unknown word, and the process of writing the i-th character, which is an unknown word, into the buffer is also a process for adding an “unknown word” flag.

  (Step S1509) The element dividing unit 1320731 increments the counter i by 1. The process returns to step S1502.

  (Step S1510) The element dividing unit 1320731 reads from the dictionary information storage unit 132070 the function word information, the pre-connection symbol, and the post-connection symbol that are paired with the first character string that matches the character string arranged on the buffer. Place on top.

  (Step S1511) The element dividing means 1320731 gives the function word information, the pre-connection symbol, and the post-connection symbol read out in step S1510 to the character string arranged on the buffer.

  (Step S1512) The element dividing unit 1320731 substitutes “i + the number of characters in the character string” for the counter i. The process returns to step S1502. Note that the “character string” of the “number of characters in the character string” is a character string arranged on the buffer.

  (Step S1513) The element connecting unit 1320732 connects character strings that can be connected among the character strings divided by the element dividing unit 1320731. Such connection processing will be described in detail with reference to the flowchart of FIG.

  (Step S1514) The character string acquisition unit 1320733 obtains the second character string from the function word information or the post-connection symbol corresponding to the last first character string of the plurality of second character strings acquired by the element connection unit 1320732. Information for specifying the type is determined, and information for specifying the type of each of the plurality of second character strings and a plurality of pairs of information in which the plurality of second character strings are paired are arranged on the memory. Such processing (referred to as sentence division processing) will be described in detail with reference to the flowchart of FIG.

  (Step S1515) The character string acquisition unit 1320733 pairs each of the two or more divided second character strings acquired in step S1514 and information specifying the type of each second character string, and arranges them on the memory. . Such memory arrangement processing will be described in detail with reference to the flowchart of FIG. The process ends. Return to upper process.

  Next, the connection process of step S1513 will be described in detail using the flowchart of FIG.

  (Step S1601) The element connecting unit 1320732 is a continuous unknown word to which the element dividing unit 1320731 has an unknown word flag (a kind of unknown word information), and forms a character string by connecting hiragana characters. The element connecting unit 1320732 is a continuous unknown word to which the element dividing unit 1320731 has an unknown word flag, and connects non-Hiragana characters to form a character string.

  (Step S1602) The element linking means 1320732 gives collective information to the character strings linked in step S1601 or the characters of unknown words that are not linked. The collective information is information that identifies the type of character string indicating the collective. Collective information is a type of function word information. The group information may be an unknown word flag added to the character string.

  (Step S1603) The element coupling means 1320732 substitutes 1 for the counter i.

  (Step S1604) The element linking means 1320732 determines whether or not the (i + 1) th combined character string (which may be one character) exists in the sentence to be processed. If the (i + 1) th character string exists, the process goes to step S1605, and if the (i + 1) th character string does not exist, the process returns to the upper process.

  (Step S1605) The element coupling means 1320732 reads the i-th character string post-connection symbol and the (i + 1) -th character string pre-connection symbol, and arranges them on the memory. The i-th character string post-connection symbol is a post-connection symbol paired with the last first character string of the i-th character string. The pre-connection symbol of the (i + 1) th character string is a pre-connection symbol paired with the first character string at the head of the (i + 1) th character string.

  (Step S1606) The element connecting means 1320732 calculates the i-th character string and the (i + 1) -th character from the post-connection symbol of the i-th character string read in Step S1605 and the pre-connection symbol of the (i + 1) -th character string. It is determined whether or not character strings can be connected. Whether or not two consecutive character strings are connectable can be determined from the preceding connection symbol and the preceding connection symbol described above. The element connecting means 1320732 can be connected, for example, 2 It holds information (rules) of a set of a back connection symbol and a front connection symbol of two character strings.

  (Step S1607) The element connecting means 1320732 goes to step S1608 if the determination result in step S1606 is that connection is possible, and goes to step S1609 if it is determined that connection is not possible.

  (Step S1608) The element connecting means 1320732 connects the i-th character string and the (i + 1) -th character string. The concatenation of character strings may be a process of writing the concatenated character strings into a buffer or a process of linking two character strings.

  (Step S1609) The element connecting means 1320732 increments the counter i by 1. Go to step S1604.

  Next, the sentence division process in step S1514 will be described in detail with reference to the flowchart in FIG.

  (Step S1701) The character string acquisition unit 1320733 substitutes 1 for the counter i.

  (Step S1702) The character string acquisition unit 1320733 determines whether or not the i-th second character string exists in the sentence to be processed. If the i-th second character string exists, the process goes to step S1703, and if the i-th second character string does not exist, the process returns to the upper process.

  (Step S1703) The character string acquisition means 1320733 determines information specifying the type of each second character string from the function word information or the post-connection symbol corresponding to the last first character string of the i-th second character string. To do. Specifically, for example, the type of the second character string is, for example, “subject”, “subtitle”, “action”, or “state”, and corresponds to the first character string at the end of the second character string. If the function word information is “operation” or “state”, the character string acquisition unit 1320733 determines that the type of the second character string is “operation” or “state”. In addition, when the subsequent connection symbol corresponding to the first character string at the end of the second character string is a symbol indicating “subject” or “subtitle”, the type of the second character string is “subject” or “subtitle”. If there is, the character string acquisition unit 1320733 determines. Then, the character string acquisition unit 1320733 takes out the i-th second character string and information specifying the type of the second character string, and arranges them on the memory.

  (Step S1704) The character string acquisition unit 1320733 increments the counter i by 1. The process returns to step S1702.

  Next, the memory arrangement processing in step S1515 will be described in detail with reference to the flowchart in FIG.

  (Step S1801) The character string acquisition unit 1320733 acquires information of the sentence structure to be output. The sentence structure information to be output is stored in advance, for example, by the character string acquisition unit 1320733, and the character string acquisition unit 1320733 reads this information. The sentence structure information to be output is, for example, “subject, subtitle, action | state”. This indicates that “character string indicating the subject”, “character string indicating the subtitle”, “character string indicating the operation”, or “character string indicating the state” are output in this order. “Subject”, “Subtitle”, “Action”, “State”, and the like indicate that the function or significance of the character string in the sentence is indicated. Further, the “subject”, “subtitle”, “motion”, “state”, and the like are hereinafter referred to as sentence elements as appropriate.

  (Step S1802) The character string acquisition unit 1320733 substitutes 1 for the counter i.

  (Step S1803) The character string acquisition unit 1320733 determines whether or not the i-th sentence element exists in the sentence structure information read in step S1801. If the i-th sentence element exists, the process goes to step S1804, and if the i-th sentence element does not exist, the process returns to the upper process.

  (Step S1804) The character string acquisition unit 1320733 reads information on the i-th sentence element (for example, “subject”) and searches for a second character string indicating the type (for example, “subject”) corresponding to the sentence element. To do.

  (Step S1805) The character string acquisition unit 1320733 goes to step S1806 if there is a second character string indicating the type (for example, “subject”) corresponding to the sentence element, and goes to step S1807 if it does not exist.

  (Step S1806) The character string acquisition unit 1320733 acquires a second character string indicating a type (for example, “subject”) corresponding to the sentence element, and pairs the second character string and the type corresponding to the sentence element. To add to the buffer.

  (Step S1807) The character string acquisition unit 1320733 increments the counter i by one. The process returns to step S1803.

  Next, details of the score change processing in step S1407 will be described using the flowchart of FIG.

  (Step S1901) The change word acquisition unit 132077 uses the change word information included in the change information of the change information storage unit 132072 to determine whether there is a change word within a predetermined number of characters before and after the evaluation word. Determine whether.

  (Step S1902) The changed word acquisition unit 132077 goes to step S1903 if it is determined that there is a changed word in step S1901, and returns to the higher-level processing if it is determined that there is no changed word. If you return here, the score has not changed.

  (Step S1903) The change word acquisition unit 132077 acquires all the character strings within a predetermined number of characters before and after the evaluation word that matches the change word included in the change information of the change information storage unit 132072, and arranges it on the memory To do.

  (Step S1904) The evaluation change information acquisition unit 132078 acquires from the change information storage unit 132072 each evaluation change information paired with one or more change words acquired in step S1903.

  (Step S1905) The modified score calculation unit 13209 changes the score using the one or more pieces of evaluation change information acquired in step S1904 with respect to the score acquired by the score acquisition unit 1320761, and calculates a new score. The change score calculation means 13209 calculates a new score by, for example, “score × first evaluation change information × second evaluation change information ×... × nth evaluation change information”. Return to upper process. At that time, if there is a changed word, a new score is returned, or if there is no changed word, the original score is returned.

  Hereinafter, a specific operation of the advertisement output system in the present embodiment will be described. FIG. 1 is a conceptual diagram of the advertisement output system.

  Now, the text storage unit 1202 of the server device 12 stores a text management table as shown in FIG. The character storage unit 1203 stores the character management table shown in FIG.

  The dictionary information storage means 132070 holds the dictionary information management table shown in FIG. The dictionary information management table stores one or more records having attributes of “first character string”, “previous connection symbol”, “rear connection symbol”, and “function word information”.

  “-&” In the “previous connection symbol” in FIG. 20 indicates that an unknown word previously identified by unknown word information, an object identified by “object”, and the like are connected. The “pre-connection symbol” also includes “&&”, etc., and “&&” indicates that “name”, unknown word, object identified by “object”, etc. are linked before. Show. “Name” is the name of a registered person or object.

  In addition, “s%” of the “rear connection symbol” indicates that it can be a delimiter and can be a “subtitle”. Further, “&&” in the “rear connection symbol” indicates that “name”, an unknown word identified by unknown word information, an object identified by “object”, and the like are connected later. In addition, “m%” of the “rear connection symbol” can be a delimiter and can be a “subject”. In addition, “n%” of the “rear connection symbol” indicates that it can be a delimiter and can be a “subject”. “M%” and “n%” mean that both can be “subjects”, but “m%” has a higher priority of becoming a subject. In addition, “-” in the “rear connection symbol” indicates that a character string identified by “action” is connected. In addition, “% =” in the “rear connection symbol” indicates that it can be a delimiter and that an object identified by “object”, “name”, and the like are connected. In addition, “o%” of “post-connection symbol” indicates that it can be a delimiter and can be “quasi-theme”. In addition, “%%” of the “rear connection symbol” indicates a forced separation.

  In addition, “object” of “function word information” indicates an object. Further, “action” of “function word information” indicates an operation. Further, “state” of “function word information” indicates a state.

  The evaluation information storage unit 132071 stores the evaluation information management table shown in FIG. The evaluation information management table stores one or more records (evaluation information) having attribute values of “evaluation word” and “score”. In this evaluation information management table, for example, the score of the evaluation word “good” is “+3”, the score of the evaluation word “fashionable” is “+4”, and the like.

  The change information storage unit 132072 stores the change information management table shown in FIG. The change information management table stores one or more records (change information) having attribute values of “change word” and “evaluation change information”. In this evaluation information management table, “evaluation change information” is a numerical value multiplied by the score.

  Further, the advertisement information storage unit 13201 stores an advertisement information management table as shown in FIG. The advertisement information management table manages a plurality of records having attribute values of “ID”, “object”, “condition”, and “advertisement information”. The “object” is an attribute that can store one or more pieces of information related to the advertisement target. “Condition” is an attribute that can store a condition for outputting advertisement information. “Advertisement information” is an attribute that can store one or more file names storing advertisement information. The “advertisement information” may store the advertisement information itself. Note that the advertisement information storage unit 13201 also holds the actual state of the file indicated by “advertisement information”. The advertisement information may have any data structure such as a character string, a still image, a moving image, a sound, and a combination of a moving image and sound. In the record of “ID = 1” in the advertisement information management table of FIG. 23, when it is determined that the user has an affirmative degree of score “+3” or higher for “cat”, “cat” or “cat” , Indicating that a still image of the advertisement information “cat goods.gif” is output to the information terminal device 11 of the user.

  In such a situation, the user of the information terminal device 11 enjoys chatting with the virtual character of the server device 132. Then, the user inputs a sentence such as “I like cats”. Such processing is as described in the first embodiment. Then, it is assumed that the user inputs the following text before pressing the “end” button in FIG.

  That is, it is assumed that the user inputs “Store X is fashionable. Store X has a reputation on television” during a conversation with a virtual character. Next, the text receiving unit 1105 receives the text “Store X is fashionable. Store X has a reputation on television”. Next, the sentence output unit 1108 outputs the accepted sentence to the “Your Remarks” field. Then, the terminal-side text transmission unit 1106 transmits the received text “Store X is fashionable. Store X has a reputation on television”.

  Next, the server-side text receiving unit 1206 receives the text “Store X is fashionable. Store X has a reputation on television” from the information terminal device 11. Next, the natural language processing unit 1207 divides the sentence into the sentence “Store X is fashionable” and the sentence “Store X is popular on television” and places the two sentences on the memory.

  Next, the natural language processing unit 13207 performs language processing on the sentence “Store X is fashionable” as follows, and obtains a processing result. First, the sentence division control unit 132073 of the natural language processing unit 13207 divides each read sentence as follows. First, the sentence division control unit 132073 performs the following division process on the sentence “Store X is fashionable”. The element division unit 1320731 constituting the sentence division control unit 132073 is the sixth character from the first character (the above “n” is “5”) in the sentence “Store X is fashionable”. The character string “Store X is o” is acquired and placed on the buffer ((1) in FIG. 24).

  The element dividing unit 1320731 then searches the dictionary information storage unit 132070 for the character string “Store X is o”. However, since the first character string that matches the character string “Store X is o” does not exist in the dictionary information management table of FIG. 20, the element dividing unit 1320731 has the character string “Store X less than the previous character string. Constitutes "". Then, the element dividing unit 1320731 places the character string “Store X is” on the buffer ((2) in FIG. 24). Similarly, the element dividing unit 1320731 searches the dictionary information management table for the character string “Store X is”. However, since the first character string that matches the character string “Store X is” does not exist in the dictionary information management table of FIG. 20, the element dividing unit 1320731 has a character string “Store X is less than the previous character string. Is configured and placed on the buffer ((3) in FIG. 24). Then, the element dividing unit 1320731 searches the dictionary information storage unit 132070 for the character string “Store X is”. Similarly, the first character string that matches the character string “Store X is” does not exist in the dictionary information management table of FIG. The element dividing unit 1320731 then searches the dictionary information storage unit 132070 by reducing the number of characters one character at a time through the same process ((4) to (6) in FIG. 24). Then, the element dividing unit 1320731 determines that the last character “Store” does not exist in the dictionary information management table of FIG. Then, the element dividing unit 1320731 gives an “unknown word” flag, which is information indicating an unknown word, to the character “store” ((7) in FIG. 24). The “unknown word” flag here is information “unknown word”.

  Next, the element dividing unit 1320731 shifts the first pointer of the character string to be processed by one. Then, the element dividing unit 1320731 reads out the character string “Pho X is oshi” for six characters from the second character and places it in the buffer ((8) in FIG. 24). Similarly, the element dividing unit 1320731 searches the dictionary information storage unit 132070 for the character string “Pho X is Oshi”. Similarly, the first character string that matches the character string “X is a letter” does not exist in the dictionary information management table of FIG. The element dividing unit 1320731 then searches the dictionary information storage unit 132070 by reducing the number of characters one by one by the same processing ((9) to (13) in FIG. 24). Then, the element dividing unit 1320731 determines that the last character “po” does not exist in the dictionary information management table of FIG. Then, the element dividing unit 1320731 adds an “unknown word” flag, which is information indicating an unknown word, to the character “po” ((14) in FIG. 24).

  Next, the element dividing unit 1320731 shifts the first pointer of the character string to be processed by one. Then, the element dividing unit 1320731 reads out the character string “X is osha” for six characters from the third character and arranges it in the buffer ((15) in FIG. 24). Similarly, the element dividing unit 1320731 searches the dictionary information storage unit 132070 for the character string “X is Osha”. Similarly, the first character string that matches the character string “X is osha” does not exist in the dictionary information management table of FIG. The element dividing unit 1320731 then searches the dictionary information storage unit 132070 by reducing the number of characters one by one by the same processing ((9) to (13) in FIG. 24). Then, the element dividing unit 1320731 determines that the last character “X” does not exist in the dictionary information management table of FIG. Then, the element dividing unit 1320731 adds an “unknown word” flag, which is information indicating an unknown word, to the character “X” ((21) in FIG. 24).

  Next, the element dividing unit 1320731 shifts the first pointer of the character string to be processed by one. Then, the element dividing unit 1320731 reads the character string “has fashion” for six characters from the fourth character and places it in the buffer ((22) in FIG. 24). Similarly, the first character string that matches the character string “has fashionable” does not exist in the dictionary information management table of FIG. The element dividing unit 1320731 then searches the dictionary information storage unit 132070 by reducing the number of characters one by one by the same process ((23) to (27) in FIG. 24). The element dividing unit 1320731 determines that the last character “ha” exists in the dictionary information management table of FIG. The element dividing unit 1320731 then forms the front connection symbol “-&”, the rear connection symbol “m%”, and the function word information “object” paired with the first character string “ha” that matches the character string arranged on the buffer. Is read from the dictionary information storage means 132070 and placed on the memory. Then, the element dividing unit 1320731 gives the read previous connection symbol “-&”, the subsequent connection symbol “m%”, and the function word information “object” to the character string “ha” arranged on the buffer. ((28) in FIG. 24).

  The above processing is repeated, and data such as (x) in FIG. 24 is obtained for the sentence “Store X is fashionable”. That is, “unknown word” is assigned to “store”, “po”, “X”, “o”, “shi”, “nya”, and “re”. Further, “− &: m%: object” is given to “ha”. Also, “: %%:” is given to “,” “.”. “: %%:” indicates that the post-connection symbol is “%%”. The back connection symbol “%%” indicates that the back is unconditionally separated. Furthermore, “− &:% −: state” is assigned to “da” ((x) in FIG. 24).

  Next, the element connecting unit 1320732 is a continuous unknown word to which the element dividing unit 1320731 has added the unknown word flag “unknown word”, and connects the characters “store”, “po”, and “X” that are not hiragana characters to the character string “ Store X ". The element connecting means 1320732 is a continuous unknown word to which the element dividing means 1320731 is assigned an unknown word flag “unknown word”, and the continuous characters “o”, “shi”, “nya”, and “re” of hiragana are connected. Constructs the string “fashionable”. Then, the element connecting unit 1320732 is a piece of information (here, “unknown word”) that identifies the type of character string that indicates the character string that is connected or the character of the unknown word that is not connected. ]) And write to the buffer. The information in the buffer is (x + 1) in FIG.

  Next, the element connecting unit 1320732 determines whether or not the character strings (for example, “I” and “to”) before and after each of FIG. 24 (x + 1) are connected, and those to be connected are connected.

  For example, for “store X” and “ha”, “store X” is an unknown word, and the element connecting means 1320732 determines to connect by “− &” before the connection symbol “−”. This is because the previous connection symbol “-&” is connected to the previous unknown word. Then, the element connecting unit 1320732 gives “: m%: object” to “Store X is” (see FIG. 24 (x + 2)). “: M%: object” indicates that there is no front connection symbol, the rear connection symbol is “m%”, and the function word information is “object”.

  The element connecting unit 1320732 determines that “Store X is” and “,” are not connected. “,” Is a symbol that is unconditionally delimited and is not combined with the preceding character string.

  Then, the element connecting means 1320732 determines that “fashionable” and “da” are connected, and connects them. That is, the element connecting means 1320732 forms the character string “fashionable” and assigns “:% −: state” (see FIG. 24 (x + 2)).

  Then, the element connecting unit 1320732 determines that “fashionable” and “.” Are not connected. “.” Is a symbol that is unconditionally delimited and is not combined with the preceding character string.

  Through the above processing, the element connecting means 1320732 obtains two second character strings (see FIG. 24 (x + 2)).

  Next, the character string acquisition unit 1320733 acquires the first second character string “Store X is”, and the function word information “object” corresponding to the last first character string “Ha” of the second character string. Alternatively, it is determined that the type of the second character string is “subject” from the post-connection symbol “m%”. Here, the character string acquisition means 1320733 determines that the type of the second character string “Store X is” is “Subject” from the post-connection symbol “m%” ((x + 3) in FIG. 24). .

  Next, the character string acquisition means 1320733 acquires the second second character string “fashionable”, and the function word information “state” corresponding to the last first character string “DA” of the second character string. From the subsequent connection symbol “%-”, the type of the second character string is determined to be “state”. Here, the character string obtaining unit 1320733 determines from the function word information “state” that the type of the second character string “fashionable” is “state” ((x + 3) in FIG. 24).

  Next, the character string acquisition unit 1320733 reads the sentence structure “subject, state” indicating the order of the elements to be output (arranged on the memory). It is assumed that the sentence structure “subject, state” is held in advance by the character string acquisition unit 1320733. Then, the character string acquisition unit 1320733 arranges the second character string “Store X is” and “It is fashionable” on the memory in accordance with the sentence structure “subject, state”. Then, the character string acquisition unit 1320733 writes the type of information corresponding to the sentence element in the memory in a pair with the second character string. Then, the character string acquisition unit 1320733 obtains “Store X: Theme, Fashionable: State”. With the above processing, the sentence “Store X is fashionable” can be divided into “theme” and “state”.

  Next, the evaluation object acquisition means 132074 uses the second character string “Store X is included in the second character string including the evaluation object from the type“ theme ”of the second character string acquired by the sentence division control means 132073 from the information arranged on the memory. ”Is specified, and the evaluation target“ store X ”is acquired from the second character string“ store X is ”.

  Next, the score acquisition unit 1320761 uses the second character string “corresponding to the type (for example,“ state ”) of another specific second character string in the sentence having the second character string“ Store X ”including the evaluation target. Get fashionable.

  Next, the score acquisition unit 1320761 reads one or more evaluation words that the evaluation information storage unit 132071 has in order to determine whether or not the evaluation word is included in the acquired second character string “fashionable”, Using the evaluation words as keys, the second character string “fashionable” is searched. Then, it is detected that one evaluation word “fashionable” in the evaluation information storage unit 132071 is included in the second character string “fashionable”.

  Then, the score acquisition unit 1320761 searches the evaluation information storage unit 132071, reads the score “+4” corresponding to the evaluation word “fashionable”, and arranges it on the memory.

  Next, an attempt is made to perform a score change process, but since the changed word is not included in the target sentence, the score change process is not performed.

  Next, the evaluation calculation means 1320762 uses the one or more scores for the same evaluation object acquired by the evaluation object acquisition means 132074 from the remarks of the same user in the past to evaluate the evaluation object (score). Is calculated. However, since the sentence “Store X is fashionable. Store X is popular on television” is the first remark made by the user regarding store X, the evaluation calculation means 1320762 performs cumulative score calculation processing. The score “+4” is stored in the memory as it is.

  Next, the sentence division control unit 132073 processes the second sentence “Store X has a reputation on television” in the same manner as described above. That is, the element division unit 1320731 of the sentence division control unit 132073 is the sixth character from the first character in the sentence “Store X is popular on television” (the above “n” is “5”). The character string “store X is te” up to the character is acquired and placed on the buffer ((1) in FIG. 25).

  The element dividing unit 1320731 then searches the dictionary information storage unit 132070 for the character string “Store X is Te”. However, since the first character string that matches the character string “store X is te” does not exist in the dictionary information management table of FIG. 20, the element dividing unit 1320731 has a character string “store X that is one character less than the previous character string. Is configured. Then, the element dividing unit 1320731 places the character string “Store X is” on the buffer ((2) in FIG. 25). Similarly, the element dividing unit 1320731 searches the dictionary information storage unit 132070 for the character string “Store X is”. However, since the first character string that matches the character string “Store X is” does not exist in the dictionary information management table of FIG. 20, the element dividing unit 1320731 has a character string “Store X is one character less than the previous character string. Is placed on the buffer ((3) in FIG. 25). Then, the element dividing unit 1320731 searches the dictionary information storage unit 132070 for the character string “Store Xga”. Similarly, the first character string that matches the character string “Store X is” does not exist in the dictionary information management table of FIG. The element dividing unit 1320731 then searches the dictionary information storage unit 132070 by reducing the number of characters one character at a time by the same processing ((4) to (6) in FIG. 25). Then, the element dividing unit 1320731 determines that the last character “Store” does not exist in the dictionary information management table of FIG. Then, the element dividing unit 1320731 adds an “unknown word” flag, which is information indicating an unknown word, to the character “store” ((7) in FIG. 25). The “unknown word” flag here is information “unknown word”.

  Next, the element dividing unit 1320731 shifts the first pointer of the character string to be processed by one. Then, the element dividing unit 1320731 reads the character string “Pho X is tele” for six characters from the second character, and places it in the buffer ((8) in FIG. 25). Similarly, the element dividing unit 1320731 searches the dictionary information storage unit 132070 for the character string “X is tele”. Similarly, the first character string that matches the character string “X is tele” does not exist in the dictionary information management table of FIG. The element dividing unit 1320731 then searches the dictionary information storage unit 132070 by reducing the number of characters one by one by the same process ((9) to (13) in FIG. 25). Then, the element dividing unit 1320731 determines that the last character “po” does not exist in the dictionary information management table of FIG. Then, the element dividing unit 1320731 adds an “unknown word” flag, which is information indicating an unknown word, to the character “po” ((14) in FIG. 25).

  Next, the element dividing unit 1320731 shifts the first pointer of the character string to be processed by one. Then, the element dividing unit 1320731 reads out the character string “X is television” for six characters from the third character and arranges it in the buffer ((15) in FIG. 25). Similarly, the element dividing unit 1320731 searches the dictionary information storage unit 132070 for the character string “X is TV”. Similarly, the first character string that matches the character string “X is television” does not exist in the dictionary information management table of FIG. The element dividing unit 1320731 then searches the dictionary information storage unit 132070 by reducing the number of characters one by one by the same process ((9) to (13) in FIG. 25). Then, the element dividing unit 1320731 determines that the last character “X” does not exist in the dictionary information management table of FIG. Then, the element dividing unit 1320731 adds an “unknown word” flag, which is information indicating an unknown word, to the character “X” ((21) in FIG. 25).

  Next, the element dividing unit 1320731 shifts the first pointer of the character string to be processed by one. Then, the element dividing unit 1320731 reads the character string “has on TV” for six characters from the fourth character, and places it in the buffer ((22) in FIG. 25). Similarly, the first character string that matches the character string “has on TV” does not exist in the dictionary information management table of FIG. The element dividing unit 1320731 then searches the dictionary information storage unit 132070 by reducing the number of characters one character at a time through the same processing ((23) to (27) in FIG. 25). Then, the element dividing unit 1320731 determines that the last character “ga” exists in the dictionary information management table of FIG. The element dividing unit 1320731 then connects the first character string “-&”, the subsequent character “n%”, and the function word information “object”, which are paired with the first character string “GA” that matches the character string arranged on the buffer. Is read from the dictionary information storage means 132070 and placed on the memory. Then, the element dividing unit 1320731 gives the read front connection symbol “-&”, the rear connection symbol “n%”, and the function word information “object” to the character string “GA” arranged on the buffer. ((28) in FIG. 25).

  The above processing is repeated, and data such as (x) in FIG. 25 is obtained for the sentence to be processed “Store X has a reputation on television. That is, “unknown word” is assigned to “store”, “po”, “X”, “te”, “le”, “bi”, “review”, and “size”. Also, “-&: n%: object” is assigned to “ga”. Also, “: %%:” is given to “,” “.”. Further, “− &: s%: object” is given to “de”. Furthermore, “− &:% −: state” is assigned to “da” ((x) in FIG. 25).

  Next, the element connecting unit 1320732 is a continuous unknown word to which the element dividing unit 1320731 has added the unknown word flag “unknown word”, and connects the characters “store”, “po”, and “X” that are not hiragana characters to the character string “ Store X ". The element connecting unit 1320732 is a continuous unknown word to which the element dividing unit 1320731 is assigned an unknown word flag “unknown word”, and connects the characters “te”, “le”, “bi”, which are not hiragana characters, and the character string “TV” Is configured. The element connecting means 1320732 is a continuous unknown word to which the element dividing means 1320731 is assigned the unknown word flag “unknown word”, and forms the character string “reputation” by connecting characters “reputation” and “size” that are not hiragana. To do. Then, the element connecting unit 1320732 is a piece of information (here, “unknown word”) that identifies the type of character string that indicates the character string that is connected or the character of the unknown word that is not connected. ]) And write to the buffer. The information in the buffer is (x + 1) in FIG.

  Next, the element connecting means 1320732 determines whether or not the character strings (for example, “Store X” and “Ga”) before and after each of FIG. 25 (x + 1) are connected, and those to be connected are connected. .

  For example, for “store X” and “ga”, “store X” is an unknown word, and the element connection means 1320732 determines that the connection is made based on the previous connection symbol “-&” of “ga”. Then, the element coupling unit 1320732 gives “: n%: object” to “Store X is” (see FIG. 25 (x + 2)). “: N%: object” indicates that there is no preceding connection symbol, the subsequent connection symbol is “n%”, and the function word information is “object”.

  Similarly, the element connecting means 1320732 determines that “TV” and “de” are connected, and connects them. That is, the element connecting unit 1320732 constitutes the character string “on television” and assigns “: s%: object” (see FIG. 25 (x + 2)). In FIG. 25 (x + 2), “−” indicates that there is no previous connection symbol.

  Similarly, the element connecting means 1320732 determines that “reputation” and “da” are connected, and connects them. That is, the element connecting unit 1320732 forms the character string “Reputation” and assigns “:% −: state” (see FIG. 25 (x + 2)).

  Through the above processing, the element connecting means 1320732 obtains three second character strings (see FIG. 25 (x + 2)).

  Next, the character string acquisition unit 1320733 acquires the first second character string “Store Xga”, and the function word information “object” corresponding to the last first character string “GA” of the second character string. Alternatively, the type of the second character string is determined as “subject” from the post-connection symbol “n%”. Here, the character string acquisition unit 1320733 determines that the type of the second character string “Store X is” is “Subject” from the post-connection symbol “n%” ((x + 3) in FIG. 25). .

  Next, the character string acquisition unit 1320733 acquires the second second character string “on TV”, and the function word information “object” corresponding to the last first character string “de” of the second character string or From the subsequent connection symbol “s%”, the type of the second character string is determined to be “subtitle”. Here, the character string acquisition unit 1320733 determines that the type of the second character string “on TV” is “subtitle” from the post-connection symbol “s%” ((x + 3) in FIG. 25).

  Next, the character string acquisition unit 1320733 acquires the third second character string “Reputation”, and the function word information “state” corresponding to the last first character string “DA” of the second character string. From the subsequent connection symbol “%-”, the type of the second character string is determined to be “state”. Here, the character string obtaining unit 1320733 determines from the function word information “state” that the type of the second character string “reputation” is “state” ((x + 3) in FIG. 17).

  Next, the character string acquisition unit 1320733 reads a sentence structure “subject, subtitle, state” indicating the order of elements to be arranged on the memory. It is assumed that the sentence structure “subject, subtitle, state” is held in advance by the character string acquisition unit 1320733. Then, the character string acquisition unit 1320733 arranges the second character strings “Store X is”, “On TV”, and “Reputation” in the memory in accordance with the sentence structure “subject, subtitle, state”. Then, the character string acquisition unit 1320733 writes the type of information corresponding to the sentence element in the memory in a pair with the second character string. Then, the character string obtaining unit 1320733 obtains “Store X: Theme, TV: Subtitle, Reputation: State”. With the above processing, the sentence “Store X is popular on television” can be divided into “subject”, “subtitle”, and “state”.

  Next, the evaluation target acquisition unit 132074 uses the second character string “store X is included in the second character string including the evaluation target from the type“ subject ”of the second character string acquired by the sentence division control unit 132073 from the information arranged on the memory. ”Is specified, and the evaluation target“ store X ”is acquired from the second character string“ store X is ”. Note that the evaluation target “store X” acquired in the first sentence may be used as it is, or the evaluation target “store X” may be given in advance.

  Next, the score acquisition unit 1320761 uses the second character string “corresponding to the type (for example,“ state ”) of another specific second character string in the sentence having the second character string“ Store X ”including the evaluation target. Get “Reputation”.

  Next, the score acquisition unit 1320761 reads one or more evaluation words of the evaluation information storage unit 132071 in order to determine whether or not the evaluation word is included in the acquired second character string “reputation”. The second character string “Reputation” is searched using each evaluation word as a key. Then, since there is no evaluation word stored in the evaluation information storage unit 132071, the processing of the second sentence “Store X has a reputation on television” ends.

  Next, the processing result configuration unit 132076 uses the evaluation target “store X” acquired by the evaluation target acquisition unit 132074 and the score “+4” calculated by the evaluation calculation unit 1320762 to obtain the processing result “store X, +4”. Constitute. The processing result “store X, +4” is positive preference information for the object “store X”.

  Next, the sentence acquisition unit 1208 determines whether or not preference information indicating a positive emotion exists in the processing result that is the result of the natural language processing in the natural language processing unit 1207. Here, since the sentence acquisition unit 1208 detects “+4” of the processing result “store X, +4” and the score is “+”, there is preference information indicating a positive emotion in the processing result. to decide. Then, the sentence acquisition unit 1208 acquires the object “store X” in the preference information and arranges it on the memory. Next, the sentence acquisition unit 1208 acquires the sentence “<object>! Like ~” corresponding to the preference information from the sentence storage unit 1202. Next, the sentence acquisition unit 1208 inserts the acquired object “Store X” into the acquired sentence “<Object>! Like ~” to obtain “Store X! .

  Next, the server-side text transmission unit 1209 transmits the acquired text “Store X!

  Next, the terminal-side text receiving unit 1107 receives the text “Store X! Then, the sentence output unit 1108 outputs the received sentence “Store X! Likes” to the field of “X-kun's remarks” (see 2601 in FIG. 26).

  Next, the advertisement information acquisition unit 1210 determines whether or not to acquire advertisement information using the obtained processing result “store X, +4”. Here, the advertisement information acquisition unit 1210 includes preference information, and when there is a change in emotion (change from positive emotion to negative emotion, change from negative emotion to positive emotion, normal emotion to positive emotion , Change from normal emotion to negative emotion, etc.), it is determined that the advertisement information is acquired. Then, the advertisement information acquisition unit 1210 detects that it was previously a normal emotion (score with a value of 0, an initial value) and has changed to a positive emotion (a score of +) this time. It is determined that information is acquired. Next, the advertisement information acquisition unit 1210 searches the advertisement information management table of FIG. 23 using the processing result “store X, +4”, and the record of “ID = 2” is the object “store X”, the condition It is determined that it matches ““ +4 ”or more”. Then, the advertisement information acquisition unit 1210 reads the advertisement information “Restaurant X.WMV”. Then, the advertisement information transmission unit 1211 transmits the advertisement information “Restaurant X.WMV” to the information terminal device 11.

  Next, the advertisement information receiving unit 1109 receives the advertisement information “Restaurant X.WMV” from the server device 12. Then, the advertisement information output unit 1110 outputs the advertisement information “Restaurant X.WMV” so as to be added to a predetermined advertisement area (see 2602 in FIG. 26).

  Next, the character transmission unit 1205 determines whether or not to acquire a character by using the obtained processing result “store X, +4”. That is, the character transmission unit 1205 determines from “store X, +4” that a positive term is included (mode 1), and obtains a character because a change has occurred from mode 3 which is the current mode. It is determined that Then, the character transmission unit 1205 acquires the character “laughing dog.WMV” in mode 1 corresponding to the processing result “store X, +4” from the character management table. Then, the character transmission unit 1205 transmits the acquired character “laughing dog.WMV” to the information terminal device 11.

  Next, the character receiving unit 1103 receives the character “laughing dog.WMV”. Then, the character output unit 1104 outputs the received character “laughing dog.WMV” (see 1203 in FIG. 26).

  Next, for example, it is assumed that the user inputs a sentence “Store X is not very beautiful”.

  Then, the text reception unit 1105 receives the text “Store X is not very beautiful”. Next, the sentence output unit 1108 outputs the accepted sentence to the “Your Remarks” field. Then, the terminal-side text transmission unit 1106 transmits the received text “Store X is not very beautiful”.

  Next, the server-side text receiving unit 1206 receives the text “Store X is not very beautiful” from the information terminal device 11. Next, since the natural language processing unit 1207 does not need to divide the sentence, the received sentence “Store X is not very beautiful” is arranged in the memory as it is.

  Next, the natural language processing unit 13207 processes the sentence “Store X is not very beautiful” as follows, and obtains a processing result.

  That is, the sentence division control unit 132073 divides each read sentence as follows. First, the sentence division control unit 132073 performs the division process described above for the sentence “Store X is not very beautiful”. Then, the sentence division control unit 132073 divides the sentence “Store X is not very beautiful” as shown in (1) of FIG. 27, and the second character string “Store X” which is the “subject”. ”And the second character string“ not very beautiful ”which is“ state ”.

  Next, the evaluation object acquisition means 132074 uses the second character string “Store X is included in the second character string including the evaluation object from the type“ theme ”of the second character string acquired by the sentence division control means 132073 from the information arranged on the memory. ”Is specified, and the evaluation target“ store X ”is acquired from the second character string“ store X is ”.

  Next, the score acquisition unit 1320761 uses the second character string “corresponding to the type (for example,“ state ”) of another specific second character string in the sentence having the second character string“ Store X ”including the evaluation target. "Not very beautiful".

  Next, the score acquisition unit 1320761 uses one or more evaluation words of the evaluation information storage unit 132071 to determine whether or not the evaluation word is included in the acquired second character string “not very beautiful”. Read and search for the second character string “not very beautiful” using each evaluation word as a key. Then, it is detected that one evaluation word “beautiful” in the evaluation information storage unit 132071 is included in the second character string “not very beautiful”.

  Then, the score acquisition unit 1320761 searches the evaluation information storage unit 132071, reads the score “+4” corresponding to the evaluation word “beautiful”, and arranges it on the memory.

  Next, score change processing is performed. That is, the change word acquisition unit 132077 uses the change word information included in the change information of the change information storage unit 132072 to set the number of characters within a predetermined number before and after the evaluation word “beautiful” (in this case, for example, “5”). .)) To determine whether there is a change word within the distance. The change word acquisition means 132077 detects that there are two change words “not so” and “not” within 5 characters before and after the evaluation word “beautiful” in the second character string “not so beautiful”. The change words “too much” and “not” are acquired and placed on the memory.

  Next, the evaluation change information acquisition unit 132078 receives the evaluation change information “0.5” paired with the acquired change word “too” and the evaluation change information “−1” paired with the change word “not”. , Obtained from the change information storage means 132072 and placed on the memory.

  Next, the change score calculation means 13209 performs an operation on the score “+4” using the two pieces of evaluation change information “0.5” and “−1”. Here, for example, the modified score calculation means 13209 performs an operation of “(+4) × 0.5 × (−1) = − 2”, obtains a new score “−2”, and arranges it on the memory. The change score calculation means 13209 may calculate a new score with a predetermined calculation formula using the obtained evaluation change information as a parameter, and the calculation formula is not limited.

  Next, the evaluation calculation means 1320762 uses the one or more scores for the same evaluation object acquired by the evaluation object acquisition means 132074 from the remarks of the same user in the past to evaluate the evaluation object (score). Is calculated. That is, the evaluation calculation means 1320762 calculates the weighted average “(((+4) + (2 × (−2))) / 3) from the previous score“ +4 ”of“ Store X ”and the current score“ −2 ”. = 0 ”is calculated as the cumulative score, and the score“ 0 ”is arranged in the memory. Next, the score processing result composition unit 1320763 uses the evaluation object “store X” acquired by the evaluation object acquisition unit 132074 and the score “0” calculated by the evaluation calculation unit 1320762, and uses the processing result “store X, 0”. Configure. It is assumed that the score processing result constituting unit 1320763 also holds “store X, −2”, which is the processing result of the sentence “store X is not very beautiful”.

  Next, the sentence acquisition unit 1208 affirms the processing result “store X, -2” that is the result of the natural language processing of the sentence “Store X is not very beautiful” in the natural language processing unit 1207. It is determined whether or not there is preference information indicating a typical emotion. Here, since the score “0” is not preference information indicating a positive emotion, the sentence acquisition unit 1208 randomly acquires sentences corresponding to the terms from the sentence storage unit 1202 to form sentences. Then, the server-side sentence transmission unit 1209 transmits the acquired sentence to the information terminal device 11. Next, the terminal-side sentence receiving unit 1107 receives a sentence from the server device 12. Then, the sentence output unit 1108 outputs the received sentence to the “X-kun's remark” field.

  Next, the advertisement information acquisition unit 1210 determines whether or not to acquire advertisement information using the obtained processing result “store X, 0”. Here, the advertisement information acquisition unit 1210 includes preference information, and when there is a change in emotion (change from positive emotion to negative emotion, change from negative emotion to positive emotion, normal emotion to positive emotion , Change from normal emotion to negative emotion, etc.), it is determined that the advertisement information is acquired. Then, the advertisement information acquisition unit 1210 detects that it was previously a normal emotion (score with a value of 0, an initial value) and has changed to a positive emotion (a score of +) this time. It is determined that information is acquired. Next, the advertisement information acquisition unit 1210 searches the advertisement information management table of FIG. 23 using the processing result “store X, 0”. However, since there is no record that matches the condition, the advertisement information transmission unit 1211 , Do not send advertising information.

  Here, since the condition “score is +4 or higher” corresponding to the previously transmitted advertisement information “Restaurant X.WMV” is not satisfied, the advertisement information “Restaurant X.WMV” displayed on the information terminal device 11 is not satisfied. May be transmitted from the server device 132 to the information terminal device 11, and the information terminal device 11 may delete the displayed advertisement information “Restaurant X.WMV”. In such a case, for example, the advertisement information acquisition unit 1210 of the server device 132 detects that a condition corresponding to the advertisement information that has already been acquired and transmitted is not met, and a deletion command transmission unit (not shown) of the server device 132 is detected. Transmits a delete command (for example, “delete restaurant X.WMV”) to the information terminal device 11. Then, the delete command receiving unit (not shown) of the information terminal device 11 receives the delete command (for example, “delete restaurant X.WMV”), and the advertisement information deleting unit (not shown) is used for the delete command (for example, “delete”). The displayed advertisement information is deleted according to the restaurant X.WMV ”).

  As described above, according to the present embodiment, since natural language processing is performed using a natural language analysis technique called functional element analysis, highly accurate language processing is possible. In particular, highly accurate language processing is possible even for colloquial sentences. As a result, advertisement information that matches the user's preference can be output, and the advertising effect is greatly improved.

  In addition, according to the present embodiment, since natural language processing is accurately performed by using subsequent phrases (such as “no”), language processing with higher accuracy can be performed.

  It should be noted that according to the present embodiment, language processing that considers the following wording (such as “no”) is not essential.

  Further, according to the present embodiment, the advertisement information is held by the server device. However, the advertisement information may be held by another third device, and the advertisement information may be transmitted to the information terminal device 11 from the third device via the server device or directly from the third device. In other words, the following advertisement output method may be realized by this system. The method is an advertisement output method realized by a sentence receiving unit, a natural language processing unit, an advertisement information acquiring unit, and an advertisement information transmitting unit, and receives a sentence from an information terminal device, The natural language processing is performed on the text received in the text receiving step, the processing result including at least one term is acquired, and the processing result acquired in the natural language processing step is used to acquire the advertisement information. An advertisement output method comprising: an advertisement information acquisition step; and an advertisement information transmission step of transmitting the advertisement information acquired in the advertisement information acquisition step to the information terminal device.

  Further, in the natural language processing step of the above method, preference information that is information related to the preference of the input user of the text is acquired from the text, and advertisement information that matches the preference information is acquired in the advertisement information acquisition step. That is preferred.

Also, the first character string, function word information that is information for specifying the function or type of the first character string in the sentence, and pre-connection that is information for specifying the type of character string that can be connected before the first character string A symbol and a type of character string that can be connected after the first character string, or a symbol indicating that it is a delimiter, and information that specifies the type of the second character string in a sentence delimited by the symbol One or more dictionary information having a pair of connection symbols is stored in a storage medium,
One or more evaluation information having evaluation words, which are terms related to the quality of an evaluation target, are stored in a storage medium, and the natural language processing step of the method indicates that it is separated from sentences constituting the sentence. Take out the first character string for the post-connection symbol, use the first character string as a key, divide the sentence into two or more second character strings that are two or more character strings, and the first character string at the end of the second character string Sentence division control step of acquiring information specifying the type of the second character string from the function word information or the post-connection symbol corresponding to, and acquiring information specifying the type of the second character string and the second character string And an evaluation object acquisition step of identifying a second character string including an evaluation object from the type of the second character string acquired in the sentence division control step, and acquiring an evaluation object from the second character string, and the evaluation object Second sentence containing An evaluation word that specifies a second character string that can include an evaluation word using another second character string type in a sentence having a column, searches the storage medium for the evaluation word that the second character string has, and acquires the evaluation word It is preferable to configure the processing result using the acquisition step, the evaluation object acquired in the evaluation object acquisition step, and the evaluation word acquired in the evaluation word acquisition step.

  In addition, the evaluation word which is a term relating to the quality of the evaluation target and one or more evaluation information having a score indicating the degree of good or bad are stored, and in the processing result configuration step of the advertisement output method, the evaluation word A score acquisition step for acquiring the score to be paired with, and an evaluation calculation step for calculating an evaluation for the evaluation target using one or more scores for the same evaluation target acquired in the evaluation target acquisition step; It is preferable to include a score processing result configuration step that configures a processing result using the evaluation target acquired in the evaluation target acquisition step and the score calculated in the evaluation calculation step.

In addition, in the above method, one or more change information that is a pair of evaluation change information that is information for changing an evaluation word and a change word that is a wording word indicating an evaluation change for the evaluation word is stored. The natural language processing step includes a change word acquisition step for acquiring a change word within a predetermined number of characters from the evaluation word, an evaluation change information acquisition step for acquiring evaluation change information paired with the change word, The score calculation step further includes a change score calculation step for changing the score using the evaluation change information acquired in the evaluation change information acquisition step and calculating a new score with respect to the score acquired in the score acquisition step, and the evaluation calculation In step
Evaluation of the evaluation object using one or more scores for the same evaluation object acquired in the evaluation object acquisition step and one or more new scores for the same evaluation object calculated in the modified score calculation step It is preferable to calculate.

  Moreover, the software which implement | achieves the server apparatus in this Embodiment is the following programs. In other words, this program performs natural language processing on a computer-side server-side text receiving unit that receives text from an information terminal device and text received by the server-side text receiving unit, and generates a processing result including at least one term. Using the acquired natural language processing unit, the processing result acquired by the natural language processing unit, the advertising information acquisition unit for acquiring advertising information, and the advertising information acquired by the advertising information acquisition unit are transmitted to the information terminal device It is a program for functioning as an advertisement information transmitting unit.

  Further, in the above program, the natural language processing unit acquires preference information that is information related to the preference of the input user of the text from the text, and the advertisement information acquisition unit acquires advertisement information that matches the preference information. It is preferred to have the computer function to obtain.

  Also, the first character string, function word information that is information for specifying the function or type of the first character string in the sentence, and pre-connection that is information for specifying the type of character string that can be connected before the first character string A symbol and a type of character string that can be connected after the first character string, or a symbol indicating that it is a delimiter, and information that specifies the type of the second character string in a sentence delimited by the symbol One or more dictionary information having a pair with a connection symbol is stored in a storage medium, and one or more evaluation information having an evaluation word that is a term relating to the quality of an evaluation target is stored in the storage medium, and the program The natural language processing unit is separated from evaluation information storage means for storing one or more evaluation information having evaluation words, which are terms relating to the quality of an evaluation target, and sentences constituting the sentence Take out the first character string for the connection symbol indicating that there is, and use the first character string as a key to divide the sentence into two or more second character strings that are two or more character strings. Information identifying the type of the second character string is obtained from the function word information corresponding to the first character string or the post-connection symbol, and information identifying the type of the second character string and the second character string is obtained. A sentence division control unit; and an evaluation target acquisition unit that identifies a second character string including an evaluation target from the type of the second character string acquired by the sentence division control unit, and acquires an evaluation target from the second character string; , Specifying a second character string that can include an evaluation word by using another second character string type in a sentence having the second character string including the evaluation object, and evaluating the evaluation word that the second character string has Evaluation word acquisition means for searching and acquiring from the information storage means, and previous And evaluated for the evaluation target acquisition means has acquired, by using the evaluation words the evaluation word acquisition unit has acquired, so as to comprise a processing result configuration means constituting the processing result, it is preferable to cause the computer to function.

  In addition, one or more evaluation information having a pair of an evaluation word which is a term relating to the quality of the evaluation target and a score indicating the degree of good or bad is stored, and the processing result constituting unit is paired with the evaluation word. Score acquisition means for acquiring a score, evaluation calculation means for calculating an evaluation for the evaluation object using one or more scores for the same evaluation object acquired by the evaluation object acquisition means, and the evaluation object acquisition means It is preferable that the program is made to function as having the score processing result configuring unit that configures the processing result using the evaluation object acquired by the method and the score calculated by the evaluation calculating unit.

In addition, the natural language processing unit stores one or more change information that is a pair of evaluation change information that is a word of a wording that indicates a change of evaluation with respect to the evaluation word and information for changing the evaluation. A change word acquisition unit that acquires a change word within a predetermined number of characters from the evaluation word, an evaluation change information acquisition unit that acquires evaluation change information that is paired with the change word, and the score acquisition unit With respect to the acquired score, the evaluation change information acquired by the evaluation change information acquisition means is used to further change the score and to calculate a new score, the evaluation calculation means, Using one or more scores for the same evaluation object acquired by the evaluation object acquisition means and one or more new scores for the same evaluation object calculated by the modified score calculation means , As for calculating the evaluation of the evaluation, it is preferred that the functioning of the above program.
(Embodiment 3)

  In the chat system according to the present embodiment, an advertisement output system that receives text input to a terminal and transmits an advertisement to each terminal in real time will be described. An advertisement output system that acquires and outputs advertisement information that matches a category corresponding to a term acquired from a sentence input by a user will be described. FIG. 1 is a conceptual diagram of an advertisement output system in the present embodiment.

  FIG. 28 is a block diagram of the advertisement output system in the present embodiment. The advertisement output system includes an information terminal device 281, a server device 282, and a second information terminal device 283. Here, the information terminal device 281 and the second information terminal device 283 have the same configuration. Then, the user of the information terminal device 281 and the user of the second information terminal device 283 enjoy chat using the chat function of the advertisement output system.

  The information terminal device 281 includes a reception unit 1101, an instruction transmission unit 1102, a text reception unit 1105, a terminal side text transmission unit 1106, a terminal side text reception unit 1107, a text output unit 1108, an advertisement information reception unit 1109, and an advertisement information output unit 1110. It comprises.

  The server device 282 includes an advertisement information storage unit 28201, an instruction reception unit 1204, a server side sentence reception unit 28206, a natural language processing unit 13207, a server side sentence transmission unit 28209, an advertisement information acquisition unit 28210, and an advertisement information transmission unit 1211. .

  The advertisement information acquisition unit 28210 includes category conversion dictionary storage unit 282101, category acquisition unit 282102, and advertisement information acquisition unit 282103.

  The second information terminal device 283 includes a second reception unit 28301, a second instruction transmission unit 28302, a second sentence reception unit 28305, a second terminal side sentence transmission unit 28306, a second terminal side sentence reception unit 28307, and a second sentence. An output unit 28308, a second advertisement information receiving unit 28309, and a second advertisement information output unit 28310 are provided.

  The advertisement information storage unit 28201 stores advertisement information corresponding to the category. The advertisement information storage unit 28201 normally stores a plurality of sets of categories and advertisement information. The advertisement information storage unit 28201 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The server-side text receiving unit 28206 receives text (chat text) from the information terminal device 281 or the second information terminal device 283. The server-side text receiving unit 28206 is usually realized by a wireless or wired communication unit, but may be realized by a broadcast receiving unit.

  The server-side text transmission unit 28209 transmits the text received by the server-side text reception unit 28206 to the information terminal device 281 or the second information terminal device 283. The server-side text transmission unit 28209 transmits the text received from the information terminal device 281 to the second information terminal device 283, and transmits the text received from the second information terminal device 283 to the information terminal device 281. The server-side text transmission unit 28209 is usually realized by a wireless or wired communication unit, but may be realized by a broadcasting unit.

  The advertisement information acquisition unit 28210 acquires the advertisement information using the processing result acquired by the natural language processing unit 13207. The advertisement information acquisition unit 28210 preferably acquires advertisement information that matches the preference information included in the processing result. The advertisement information acquisition unit 28210 can usually be realized by an MPU, a memory, or the like. The processing procedure of the advertisement information acquisition unit 28210 is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The category conversion dictionary storage unit 282101 stores a category conversion dictionary indicating the correspondence between terms and categories. The category conversion dictionary may be a so-called concept dictionary. The category conversion dictionary storage unit 282101 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The category acquisition unit 282102 acquires from the category conversion dictionary storage unit 282101 a category corresponding to a term (a concept including the “subject” and “object”) included in the processing result acquired by the natural language processing unit 13207. . The category acquisition unit 282102 is usually realized by an MPU, a memory, or the like. The processing procedure of the category acquisition unit 282102 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The advertisement information acquisition unit 282103 acquires advertisement information corresponding to the category acquired by the category acquisition unit 282102 from the advertisement information storage unit 28201. The advertisement information acquisition unit 282103 can usually be realized by an MPU, a memory, or the like. The processing procedure of the advertisement information acquisition unit 282103 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The second reception unit 28301 of the second information terminal device 283 performs the same operation as the reception unit 1101 of the information terminal device 281. Similarly, the second instruction transmission unit 28302 is the same as the instruction transmission unit 1102, the second sentence reception unit 28305 is the same as the sentence reception unit 1105, and the second terminal side sentence transmission unit 28306 is the terminal side sentence transmission. The second terminal side text receiving unit 28307 is the same as the terminal side text receiving unit 1107, the second text output unit 28308 is the same as the text output unit 1108, and the second advertisement information receiving unit 28309 is the same as the unit 1106. Is the same as the advertisement information receiving unit 1109, and the second advertisement information output unit 28310 is the same as the advertisement information output unit 1110.

  Next, the operation of the advertisement output system will be described. First, the operation of the information terminal device 281 will be described using the flowchart of FIG. The operation of the second information terminal device 283 is the same as that of the information terminal device 281 and will not be described. In FIG. 29, processing is omitted for the same steps as those in the flowchart of FIG.

  (Step S2901) It is determined whether a means (not shown) has received an application screen (for example, an application screen described in HTML or XML) for performing a chat system. If the application screen is received, the process goes to step S2902, and if the application screen is not received, the process returns to step S2901.

  (Step S2902) A means not shown outputs the application screen received in step S2901.

  Next, the operation of the server device 282 will be described using the flowchart of FIG.

  (Step S3001) A means (not shown) reads an application screen of the chat system.

  (Step S3002) The means (not shown) transmits the application screen of the chat system read out in step S3001 to the information terminal device 281 and the second information terminal device 283. Such processing is an operation of a known chat system.

  (Step S3003) The server-side text receiving unit 28206 determines whether a text has been received from the information terminal device 281 or the second information terminal device 283. If a sentence is received, the process goes to step S3004. If no sentence is received, the process returns to step S3003.

  (Step S3004) The natural language processing unit 13207 performs natural language processing by the functional element analysis described in the second embodiment on the sentence received in step S3003.

  (Step S3005) The server-side text transmission unit 28209 transmits the text received in step S3003 to the chat partner terminal (the information terminal device 281 or the second information terminal device 283).

  (Step S3006) The advertisement information acquisition unit 28210 acquires advertisement information using the natural language processing result in step S3004. Specifically, first, the category acquisition unit 282102 acquires an object from the natural language processing result. Then, the category acquisition unit 282102 acquires the category corresponding to the object from the category conversion dictionary storage unit 282101. Next, the advertisement information acquisition unit 282103 acquires advertisement information paired with the category acquired by the category acquisition unit 282102.

  Hereinafter, a specific operation of the advertisement output system in the present embodiment will be described. FIG. 31 is a conceptual diagram of the advertisement output system. In FIG. 31, the user “Mr. A” of the information terminal device 281 and the user “Mr. B” of the second information terminal device 283 are enjoying the chat. At that time, the text input by Mr. A and the text input by Mr. B are transmitted to the server device 282, and the advertisement information is transmitted from the server device 282 to the second information terminal device 283 or the information terminal device 281 for output. Is done.

  Now, the advertisement information storage unit 28201 of the server device 12 stores the advertisement information management table shown in FIG. The advertisement information management table stores one or more records having attribute values of “ID”, “category”, and “advertisement information”. The “advertisement information” may be information indicating the location of the advertisement information, such as a URL or a URI, in addition to a still image, a moving image, text, audio, and the like.

  The category conversion dictionary storage unit 282101 stores the category conversion dictionary shown in FIG. The category conversion dictionary is information of a pair of “category” and “object (term)”, and one category is associated with one or more objects. “Category” is usually information of a superordinate concept with respect to “object (term)”.

  The dictionary information storage means 132070 holds the dictionary information management table shown in FIG.

  In such a situation, Mr. A and Mr. B enter texts at their terminals (information terminal device 281 or second information terminal device 283) and chat. Here, for example, suppose that Mr. A has input the sentence “I like cats, but I do”. Then, the natural language processing unit 13207 analyzes the sentence “I like cats, but I acquire the object“ cat ”and the evaluation word“ likes ”by the process described in the second embodiment.

  Then, the score acquisition unit 1320761 searches the evaluation information storage unit 132071, reads the score “+4” corresponding to the evaluation word “like”, and arranges it on the memory.

  Next, an attempt is made to perform a score change process, but since the changed word is not included in the target sentence, the score change process is not performed.

  Next, the evaluation calculation means 1320762 uses the one or more scores for the same evaluation object acquired by the evaluation object acquisition means 132074 from the remarks of the same user in the past to evaluate the evaluation object (score). Is calculated. However, since the sentence “I like cats” is the first remark about the cat by the user, the evaluation calculation means 1320762 does not perform the cumulative score calculation process, and the score “+4” remains in the memory. Suppose that it is stored above.

  Next, the advertisement information acquisition unit 28210 determines whether or not to acquire advertisement information using the obtained processing result “cat, +4”. Here, it is assumed that the advertisement information acquisition unit 28210 includes preference information and determines that the advertisement information is acquired when there is a change in emotion. Then, the advertisement information acquisition unit 28210 detects that it was previously a normal emotion (score with a value of 0, an initial value), and has changed to a positive emotion (a score of +) this time. It is assumed that information is acquired.

  Next, the category acquisition unit 282102 reads out the category “pet” corresponding to the object “cat” from the category conversion dictionary of FIG. Then, the advertisement information acquisition unit 282103 acquires advertisement information “http://www.pet....” That is paired with the category “pet”.

  Next, the advertisement information transmission unit 1211 transmits the advertisement information “http://www.pet....” To the information terminal device 281 of Mr. A.

  Next, the advertisement information receiving unit 1109 of the information terminal device 281 receives the advertisement information “http://www.pet....”, And the advertisement information output unit 1110 is placed at an appropriate position on the chat screen. The advertisement information “http://www.pet....” Is output. Then, while chatting, Mr. A presses the advertisement information “http://www.pet....” With the mouse, thereby changing the advertisement information “http://www.pet. You can see the home page of the corresponding pet.

  Note that the sentence input by Mr. B is processed in the same manner, and the advertisement information is similarly output to the second information terminal device 283 which is Mr. B's terminal.

  As described above, according to the present embodiment, it is possible to output advertisement information suitable for the user that matches the content of the chat to the terminal of the user who is enjoying the chat. As a result, the effect of the advertisement information is greatly improved.

  Note that the software that implements the server device in the present embodiment is the following program. In other words, this program performs natural language processing on a computer-side server-side text receiving unit that receives text from an information terminal device and text received by the server-side text receiving unit, and generates a processing result including at least one term. Using the acquired natural language processing unit, the processing result acquired by the natural language processing unit, the advertising information acquisition unit for acquiring advertising information, and the advertising information acquired by the advertising information acquisition unit are transmitted to the information terminal device It is a program for functioning as an advertisement information transmitting unit.

In addition, the advertisement information corresponding to the category is stored in the storage medium, and in the above program, the computer and the advertisement information acquiring unit store the category conversion dictionary indicating the correspondence between the term and the category. A category acquisition unit that acquires a category corresponding to a term included in a processing result acquired by the natural language processing unit from the category conversion dictionary storage unit, and an advertisement corresponding to the category acquired by the category acquisition unit A program for causing information to function as an advertisement information acquisition unit that acquires information from the storage medium.
(Embodiment 4)

  In the chat system according to the present embodiment, an advertisement output system that acquires text using dialogue linkage information and transmits it to each terminal device will be described. Dialog link information is information related to the connection of dialog sentences. FIG. 1 is a conceptual diagram of an advertisement output system in the present embodiment.

  FIG. 34 is a block diagram of a server device constituting the advertisement output system in the present embodiment. The advertisement output system includes an information terminal device 11 and a server device 342.

  The server device 342 includes an advertisement information storage unit 13201, a dialogue linkage information storage unit 34201, a text storage unit 34202, a character storage unit 1203, an instruction reception unit 1204, a character transmission unit 1205, a server side text reception unit 1206, and a natural language processing unit 13207. , A sentence acquisition unit 34208, a server side sentence transmission unit 1209, an advertisement information acquisition unit 13210, and an advertisement information transmission unit 1211.

  The text acquisition unit 34208 includes first dialog information acquisition means 320881 and dialog text acquisition means 320882.

  The dialogue linkage information storage unit 34201 stores dialogue linkage information that is information related to the linkage of dialogue sentences. The dialogue link information includes, for example, first dialogue information corresponding to the first dialogue sentence and second dialogue information corresponding to the second dialogue sentence. The first dialogue sentence is, for example, a dialogue sentence from one user, and the second dialogue sentence is a dialogue sentence from another user. In addition, the dialogue link information includes, for example, one or more second character strings corresponding to the first dialogue sentence or first dialogue information that is information composed of the one or more second character strings, and a second dialogue sentence. Second dialogue information corresponding to one or more second character strings or information composed of the one or more second character strings, and one or more second character strings corresponding to the third dialogue sentence or the one or more It has the 3rd dialog information which is the information comprised from a 2nd character string. The dialogue link information also includes, for example, priority information corresponding to the first dialogue information and priority information corresponding to the second dialogue information. The priority information is priority information when selected. Specific examples of the first dialogue information and the second dialogue information will be described later. The dialogue linkage information storage unit 34201 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The text storage unit 34202 receives a first dialogue sentence that is a dialogue sentence that constitutes the first dialogue information and a second dialogue sentence that is a dialogue sentence that constitutes the second dialogue information, respectively. The dialogue information is stored in association with the second dialogue information. A dialogue sentence is one or more sentences. A sentence is not necessarily a complete sentence. The text storage unit 34202 is preferably a non-volatile recording medium, but can also be realized by a volatile recording medium.

  The sentence acquisition unit 34208 acquires a sentence corresponding to the processing result in the natural language processing unit 13207 from the sentence storage unit 34202.

  The first dialog information acquisition unit 342081 uses the processing result in the natural language processing unit 13207 to search the dialog link information storage unit 34201 and acquires the first dialog information. The first dialogue information acquisition unit 342081 can be usually realized by an MPU, a memory, or the like. The processing procedure of the first dialogue information acquisition unit 342081 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The dialogue sentence acquisition unit 342082 acquires the second dialogue sentence corresponding to the second dialogue information paired with the first dialogue information from the sentence storage unit 34202. The dialogue sentence acquisition unit 342082 can be generally realized by an MPU, a memory, or the like. The processing procedure of the dialog sentence acquisition unit 342082 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  Next, the process of acquiring the text of the server device 342 will be described using the flowchart of FIG. Note that the processing other than the processing for acquiring the text is the same as the processing described in the second embodiment.

  (Step S3501) The first interaction information acquisition unit 342081 of the sentence acquisition unit 34208 substitutes 1 for the counter i.

  (Step S3502) The first dialogue information acquisition unit 342081 determines whether or not the i-th dialogue information pattern exists. It is assumed that the first dialogue information acquisition unit 342081 holds one or more dialogue information patterns in a storage medium (not shown). The dialogue information pattern includes, for example, “theme, state”, “theme, subtitle, state”, “the whole dialogue sentence”, “theme”, and the like. Also, priority information is stored in association with the dialogue information pattern. If the i-th dialog information pattern exists, the process proceeds to step S3503. If the i-th dialog information pattern does not exist, the process proceeds to step S3514.

  (Step S3503) The first interaction information acquisition unit 342081 acquires the i-th interaction information pattern from a storage medium (not shown).

  (Step S3504) The first dialog information acquisition unit 342081 matches the dialog information pattern acquired in Step S3504 from the text received by the server side text receiving unit 1206 or the processing result in the natural language processing unit 13207. Configure conversation information. For example, when the i-th dialog information pattern is “subject, state”, the first dialogue information acquisition unit 342081 obtains a character string corresponding to “subject” and a character string corresponding to “state” from the processing result. Obtain and place on memory.

  (Step S3505) The first dialog information acquisition unit 342081 determines whether or not the dialog information has been acquired in step S3504. If the dialog information can be acquired, the process goes to step S3505. If the dialog information cannot be acquired, the process goes to step S3513.

  (Step S3506) The first dialog information acquisition unit 342081 digitizes the dialog information acquired in Step S3504. Since the process of digitizing the dialog information is a known technique, the description thereof is omitted. Such processing is, for example, hash key generation processing. The hash key generation process is, for example, a process of folding a character string into a numerical value of a predetermined size. In addition, the hash key generation process is a process in which, for example, a character string is digitized and then divided by a predetermined value to obtain a remainder value. Further, the process of converting the dialogue information into a numerical value is simply a process of converting a character string into a numerical value.

  (Step S3507) The first interaction information acquisition unit 342081 substitutes 1 for the counter n.

  (Step S3508) The first interaction information acquisition unit 342081 determines whether there is interaction information whose priority information value is n. If n dialogue information exists, the process goes to step S3509, and if not, the process goes to step S3512.

  (Step S3509) The first dialogue information acquisition unit 342081 sends dialogue information that matches the dialogue information quantified in step S3506 among the dialogue information whose priority information value is n to the dialogue linkage information storage unit 34201. ,Search for.

  (Step S3510) The first dialogue information acquisition unit 342081 determines whether or not the dialogue information has been searched in step S3509. If the dialog information can be searched, the process goes to step S3511. If the dialog information cannot be searched, the process goes to step S3512.

  (Step S3511) The dialogue sentence acquisition unit 342082 obtains the dialogue information of the dialogue information pair obtained in Step S3510 from the dialogue linkage information storage unit 34201, and corresponds (linked) to the obtained dialogue information. A sentence is read from the sentence storage unit 34202. Return to upper process.

  (Step S3512) The first interaction information acquisition unit 342081 increments the counter n by 1. The process returns to step S3508.

  (Step S3513) The first interaction information acquisition unit 342081 increments the counter i by one. The process returns to step S3502.

  (Step S3514) The dialogue sentence acquisition unit 342082 reads a default dialogue sentence. It is assumed that the default dialog text is stored in the dialog text acquisition unit 342082. The default dialogue sentence is, for example, “Would you like to talk about“ predetermined term ”?” Or “Would you like to change the topic?”. Return to upper process. Note that <predetermined term> is a term in which a large number of corresponding sentences are stored, and is preliminarily held in the storage medium by the server device 342.

  Hereinafter, the specific operation | movement of a text acquisition process is demonstrated among operation | movement of the advertisement output system in this Embodiment. Through the text acquisition process, the user of the information terminal device 11 can interact with a virtual character on the server device 342, and an advertisement corresponding to the user's utterance is also output. In the following description, the description of the already described advertisement output process is omitted.

  Now, in the dialogue linkage information storage unit 34201 and the sentence storage unit 34202, dialogue linkage information obtained by functionally analyzing the dialogue (dialogue between Mr. A and Mr. B) in FIG. Are stored (see FIG. 37). Here, the dialogue link information includes dialogue information configured corresponding to a dialogue information pattern (for example, “subject, state”, “subject, subtitle, state”, “entire dialogue sentence”, “subject”, etc.), and priority. Information, an identifier of the speaker (for example, “A”), and an identifier of the speaker (for example, “B”). In addition, one or more pieces of dialogue information are associated with a dialogue sentence (a sentence that is a source of the dialogue information) and stored in the sentence storage unit 34202. FIG. 37 is a conceptual diagram of the dialog linkage information. Actually, for example, the dialog linkage information storage unit 34201 and the sentence storage unit 34202 are provided with the dialogue linkage index management table of FIG. 38 and the question of FIG. The data management table and the response data management table of FIG. 40 are stored.

  The dialogue linkage index management table stores one or more records having attribute values of “ID”, “immediate response number”, “question number”, and “response number”. “ID” is an ID for identifying a record. The “immediate response number”, “question number”, and “response number” are data obtained by digitizing a character string (sentence). The “immediate response number”, “inquiry number”, and “response number” indicate the connection of the dialogue. For example, the “immediate response message” that is Mr. B's utterance, and the utterance text of Mr. A that follows. This is a number corresponding to a certain “question sentence” and “response sentence” which is the next utterance sentence of Mr. B. These three attribute values indicate the connection of the dialogue. Further, in the dialogue linkage index management table, if there is Mr. A's utterance next to the response sentence indicated by “response number”, the “response number” becomes the “previous response number” of the next record, and Mr. A ’s The number corresponding to the utterance becomes the “question number” of the record.

  The inquiry data management table stores one or more records having attribute values of “inquiry number”, “object”, “priority information”, “speaker ID”, “partner ID”, and “dialog”. “Object” is dialogue information composed of the result of functional element analysis. A “dialogue sentence” is a sentence that is the basis of functional element analysis. The inquiry data management table is composed of, for example, Mr. A's remarks.

  The response data management table stores one or more records having attribute values of “response number”, “object”, “priority information”, “speaker ID”, “partner ID”, and “dialog”. The response data management table is composed of, for example, Mr. B's remarks.

  Moreover, the 1st dialog information acquisition means 342081 has stored the dialog information pattern management table | surface shown in FIG. The dialog information pattern management table stores one or more records having attribute values of “ID”, “dialog information pattern”, and “priority information”. The “dialog information pattern” is pattern information for composing the dialog information, and the “entire sentence” indicates that the text composing the dialog text is acquired as the dialog information as it is. Further, ““ subject ”,“ subtitle ”, and“ action / state ”indicate that“ subject ”,“ subtitle ”, and“ action / state ”are acquired as dialogue information from the sentence constituting the dialogue sentence. “Operation / state” indicates that either “operation” or “state” may be used. The “priority information” indicates the priority when a dialogue sentence is selected in a virtual chat system or the like. Here, it is assumed that the lower the priority, the higher the priority.

In such a situation, the user of the information terminal device 11 inputs the dialogue sentence “How about store X?”. Then, the server-side text receiving unit 1206 of the server device 342
The sentence “How about store X?” Is received. Next, the natural language processing unit 13207 performs the natural language processing on the sentence “How is the store X?” To obtain “Theme: Store X is” and “State: How”.

  Next, the first dialogue information acquisition unit 342081 acquires the first dialogue information pattern “entire sentence” from the dialogue information pattern management table of FIG. 41, and the first dialogue information corresponding to the dialogue information pattern. Information "How about store X?" (Dialog as it is) is obtained. Then, it is assumed that the first dialogue information acquisition unit 342081 digitizes the dialogue information “How is the store X?”, For example, “79273” is obtained.

  Next, the first dialogue information acquisition unit 342081 determines whether or not the dialogue information exists in order of the priority information value starting from 1. Further, the first dialogue information acquisition unit 342081 searches for dialogue information in order of the priority information value starting from 1, using “79273” as a key. Here, it is assumed that “79273” does not match for all the dialogue information.

  Next, similarly, the first dialogue information acquisition unit 342081 acquires dialogue information ““ Store X is ”” “How” (a pattern including all elements) corresponding to the second dialogue information pattern. Then, it is assumed that the first dialogue information acquisition unit 342081 digitizes the second dialogue information ““ Store X is ”” “How”, for example, “555345”.

  Next, the first dialogue information acquisition unit 342081 determines whether or not the dialogue information exists in order of the priority information value starting from 1. Further, the first dialogue information acquisition unit 342081 searches for dialogue information in order from the value of the priority information starting from 1, using “555345” as a key. Here, it is assumed that “555345” does not match for all the dialogue information.

  Next, similarly, the first dialogue information acquisition means 342081 quantifies the third dialogue information ““ how ”“ store X is ”(a pattern including all elements and changing the order of the elements), For example, it is assumed that “29001” is obtained.

  Next, the first dialogue information acquisition unit 342081 determines whether or not the dialogue information exists in order of the priority information value starting from 1. In addition, the first dialogue information acquisition unit 342081 searches for dialogue information in order of the priority information value starting from 1, using “29001” as a key. Here, it is assumed that “29001” did not match for all the dialogue information.

  Next, similarly, the first dialogue information acquisition unit 342081 numerically sets the third dialogue information ““ Store X is ”” (a pattern composed only of the subject) corresponding to the third dialogue information pattern. For example, it is assumed that “35677” is obtained.

  Next, the first dialogue information acquisition unit 342081 determines whether or not the dialogue information exists in order of the priority information value starting from 1. Further, the first dialogue information acquisition unit 342081 searches for dialogue information in order of the priority information value starting from 1, using “35677” as a key. Here, it is assumed that the dialogue information corresponding to “Store X is” in the fifth rectangle from the top in FIG.

  Next, the dialog sentence acquisition unit 342082 acquires the dialog information (dialog information in the lowest rectangle in FIG. 37) of the acquired dialog information “35677” from the dialog link information storage unit 503. Then, the dialogue sentence acquisition unit 342082 reads the dialogue sentence “Store X is popular on television” corresponding to the dialogue information from the sentence storage unit 34202.

  Then, the server-side sentence transmission unit 1209 transmits the acquired dialogue sentence “Store X has a reputation on television” to the information terminal device 11.

  Next, the information terminal device 11 receives the dialogue sentence “Store X has a reputation on television”. Then, the information terminal device 11 outputs the dialogue sentence “Store X is popular on television” to the display.

  Further, the advertisement information acquisition unit 13210 of the server device 342 acquires the advertisement information by the processing described in the above embodiment, and the advertisement information transmission unit 1211 transmits the advertisement information. And the information terminal device 11 receives and outputs advertisement information.

  As described above, according to the present embodiment, it is possible to realize a virtual chat as if the character is talking to the user. An advertisement corresponding to the user's statement is output.

  In the present embodiment, the dialogue link information may not have priority information, a user identifier, or the like.

  Further, in the present embodiment, the first interaction information acquisition unit 342081 may acquire the interaction information using the user identifier so that the user who speaks next becomes the user identified by the user identifier. . Further, the dialogue sentence acquisition unit 342082 may acquire a user's utterance sentence identified by the user identifier. In such a case, the personality of the other party of the conversation matches the personality of the user identified by the user identifier, and the user of the information terminal device 11 can enjoy the conversation with his / her favorite person, although it is virtual.

  Needless to say, the sentence acquisition process according to the present embodiment may be applied to the sentence acquisition process according to another embodiment.

  Furthermore, the software that implements the server device in the present embodiment is the following program. In other words, this program performs natural language processing on a computer-side server-side text receiving unit that receives text from an information terminal device and text received by the server-side text receiving unit, and generates a processing result including at least one term. Using the acquired natural language processing unit, the processing result acquired by the natural language processing unit, the advertising information acquisition unit for acquiring advertising information, and the advertising information acquired by the advertising information acquisition unit are transmitted to the information terminal device It is a program for functioning as an advertisement information transmitting unit.

  In addition, the storage medium stores information related to the connection of dialogue sentences, and includes dialogue linkage information having first dialogue information corresponding to the first dialogue sentence and second dialogue information corresponding to the second dialogue sentence, And a first dialogue sentence which is a dialogue sentence which constitutes the first dialogue information, and a second dialogue sentence which is a dialogue sentence which constitutes the second dialogue information are respectively designated as the first dialogue. Information and the second dialogue information are stored in association with each other. In the program, the sentence acquisition unit searches the dialogue linkage information storage unit using a processing result in the natural language processing unit, A first dialogue information obtaining unit for obtaining dialogue information; and a dialogue statement obtaining unit for obtaining a second dialogue sentence corresponding to the second dialogue information paired with the first dialogue information from the sentence storage unit. As a computer, to function Because of the program, it is.

  Further, the dialogue link information in the program includes one or more second character strings corresponding to the first dialogue sentence or first dialogue information which is information composed of the one or more second character strings, and a second dialogue. It is preferable to have one or more second character strings corresponding to a sentence or second dialogue information that is information composed of the one or more second character strings.

  In addition, it is preferable that the dialogue linkage information in the program also includes priority information corresponding to the first dialogue information and priority information corresponding to the second dialogue information.

  FIG. 42 shows the external appearance of a computer that executes the program described in this specification to realize the information terminal device and the server device of the various embodiments described above. The above-described embodiments can be realized by computer hardware and a computer program executed thereon. FIG. 42 is an overview of the computer system 340, and FIG. 43 is a block diagram of the computer system 340.

  42, a computer system 340 includes a computer 341 including an FD (Flexible Disk) drive and a CD-ROM (Compact Disk Read Only Memory) drive, a keyboard 342, a mouse 343, and a monitor 344.

  43, in addition to the FD drive 3411 and the CD-ROM drive 3412, the computer 341 includes a CPU (Central Processing Unit) 3413, a bus 3414 connected to the CPU 3413, the CD-ROM drive 3412, and the FD drive 3411, and a boot. A ROM (Read-Only Memory) 3415 for storing a program such as an up program, and a RAM (Random Access Memory) connected to the CPU 3413 for temporarily storing instructions of an application program and providing a temporary storage space 3416 and a hard disk 3417 for storing application programs, system programs, and data. Although not shown here, the computer 341 may further include a network card that provides connection to the LAN.

  A program that causes the computer system 340 to execute the functions of the information processing apparatus according to the above-described embodiment is stored in the CD-ROM 3501 or the FD 3502, inserted into the CD-ROM drive 3412 or the FD drive 3411, and further stored in the hard disk 3417. May be forwarded. Alternatively, the program may be transmitted to the computer 341 via a network (not shown) and stored in the hard disk 3417. The program is loaded into the RAM 3416 at the time of execution. The program may be loaded directly from the CD-ROM 3501, the FD 3502, or the network.

  The program does not necessarily include an operating system (OS), a third party program, or the like that causes the computer 341 to execute the functions of the server device or the like of the above-described embodiment. The program only needs to include an instruction portion that calls an appropriate function (module) in a controlled manner and obtains a desired result. How the computer system 340 operates is well known and will not be described in detail.

  In each of the above embodiments, each process (each function) may be realized by centralized processing by a single device (system), or by distributed processing by a plurality of devices. May be.

  In the above program, in the step of transmitting information, the step of receiving information, etc., processing performed by hardware, for example, processing performed by a modem or an interface card in the transmission step (only performed by hardware) Processing) is not included.

  Further, the computer that executes the program may be singular or plural. That is, centralized processing may be performed, or distributed processing may be performed.

  Further, in each of the above embodiments, it goes without saying that two or more communication means (such as an information transmission unit) existing in one apparatus may be physically realized by one medium.

  The present invention is not limited to the above-described embodiments, and various modifications are possible, and it goes without saying that these are also included in the scope of the present invention.

  As described above, the advertisement output system according to the present invention has an effect that an advertisement suitable for the user's preference or the like can be output in an interactive system such as chat, and is useful as a chat system or the like.

Conceptual diagram of advertisement output system in the present embodiment Block diagram of the advertisement output system A flowchart for explaining the operation of the information terminal device Flow chart for explaining the operation of the server device Flow chart explaining the natural language processing Flowchart explaining the sentence acquisition process Flowchart explaining the advertisement information acquisition process Figure showing the advertising information management table Figure showing the sentence management table The figure which shows the same character management table Figure showing an example of the same screen output The figure which shows the example of the same screen output Block diagram of the server device Flow chart for explaining the operation of the server device Flowchart explaining the sentence division control process Flow chart explaining the connection process Flow chart explaining the sentence division process Flow chart for explaining the memory allocation process Flowchart explaining the same score changing process The figure which shows the dictionary information management table Figure showing the same evaluation information management table Figure showing the change information management table Figure showing the advertising information management table Diagram showing the progress of the natural language processing Diagram showing the progress of the natural language processing Figure showing an example of the same screen output The figure which shows the result of the natural language processing Block diagram of the advertisement output system A flowchart for explaining the operation of the information terminal device Flow chart for explaining the operation of the server device Conceptual diagram of the advertisement output system Figure showing the advertising information management table Diagram showing the category conversion dictionary Block diagram of the server device Flow chart for explaining the operation of the server device Figure showing a specific example of the dialogue The figure which shows the example of the dialogue linkage information and sentence (dialogue sentence) stored in the same Figure showing the dialogue linkage index management table Figure showing the inquiry data management table Figure showing the response data management table Figure showing the dialogue information pattern management table External view of a computer that implements the server device etc. Block diagram of a computer system that implements the server device etc.

Explanation of symbols

11, 281 Information terminal device 12, 132, 282, 342 Server device 283 Second information terminal device 503, 34201 Dialog linkage information storage unit 1101 Reception unit 1102 Instruction transmission unit 1103 Character reception unit 1104 Character output unit 1105 Text reception unit 1106 Terminal side text transmission unit 1107 Terminal side text reception unit 1108 Text output unit 1109 Advertising information reception unit 1110 Advertising information output unit 1201, 13201, 28201 Advertising information storage unit 1202, 34202 Text storage unit 1203 Character storage unit 1204 Instruction receiving unit 1205 Character Transmission unit 1206, 28206 Server side text reception unit 1207, 13207 Natural language processing unit 1208, 34208 Text acquisition unit 1209, 28209 Server side text transmission unit 1210, 13210, 2 8210 Advertisement information acquisition unit 1211 Advertisement information transmission unit 13209 Change score calculation means 28301 Second reception unit 28302 Second instruction transmission unit 28305 Second sentence reception unit 28306 Second terminal side sentence transmission unit 28307 Second terminal side sentence reception unit 28308 Two sentence output unit 28309 Second advertisement information receiving unit 28310 Second advertisement information output unit 132070 Dictionary information storage unit 132071 Evaluation information storage unit 132072 Change information storage unit 132073 Sentence division control unit 132074 Evaluation target acquisition unit 132075 Evaluation word acquisition unit 132076 Processing Result composition means 132077 Changed word acquisition means 132078 Evaluation change information acquisition means 282101 Category conversion dictionary storage means 282102 Category acquisition means 282103 Advertisement information acquisition means 3420 81 First dialogue information acquisition means 320882 Dialogue sentence acquisition means 1320731 Element division means 1320732 Element connection means 1320733 Character string acquisition means 1320761 Score acquisition means 1320762 Evaluation calculation means 1320763 Score processing result configuration means

Claims (20)

An advertisement output system comprising an information terminal device and a server device,
The information terminal device
A sentence reception unit for receiving sentences;
A terminal-side text transmission unit that transmits the text received by the text reception unit to the server device;
In response to transmission of the text, an advertisement information receiving unit that receives advertisement information from the server device;
An advertisement information output unit for outputting the advertisement information;
The server device
A server-side text receiver that receives text from the information terminal device;
A natural language processing unit that performs natural language processing on the text received by the server side text receiving unit, and obtains a processing result including at least one term;
Using the processing result acquired by the natural language processing unit, an advertising information acquisition unit that acquires advertising information;
An advertisement output system comprising an advertisement information transmission unit that transmits the advertisement information acquired by the advertisement information acquisition unit to the information terminal device. The natural language processing unit includes:
From the sentence, obtaining preference information that is information about the preference of the input user of the sentence,
The advertisement information acquisition unit
The advertisement output system according to claim 1, wherein advertisement information that matches the preference information is acquired. The natural language processing unit includes:
A first character string, function word information that is information specifying the function or type of the first character string in the sentence, and a pre-connection symbol that is information specifying the type of character string that can be connected before the first character string; , A character string that can be connected after the first character string, or a symbol indicating that it is a delimiter, and a post-connection symbol that is information specifying the type of the second character string in the sentence delimited by the symbol Dictionary information storage means for storing one or more dictionary information having a pair of
Evaluation information storage means for storing one or more evaluation information having evaluation words, which are terms related to the quality of the evaluation object;
A first character string corresponding to a post-connection symbol indicating that the sentence is a delimiter is extracted from the sentences constituting the sentence, and the sentence is divided into two or more second character strings that are two or more character strings using the first character string as a key. , From the function word information corresponding to the last first character string of the second character string or the post-connection symbol, the information for identifying the type of the second character string is acquired, and the second character string and the second character string Sentence division control means for acquiring information identifying the type of
An evaluation object acquisition means for specifying a second character string including an evaluation object from the type of the second character string acquired by the sentence division control means, and acquiring an evaluation object from the second character string;
A second character string that can include an evaluation word is specified using another second character string type in a sentence having the second character string that includes the evaluation target, and the evaluation word that the second character string has is the evaluation information. An evaluation word acquisition means for searching and acquiring from the storage means;
The advertisement output system according to claim 2, further comprising: a processing result configuring unit that configures a processing result using the evaluation target acquired by the evaluation target acquiring unit and the evaluation word acquired by the evaluation word acquiring unit. The evaluation information storage means includes
One or more evaluation information having a pair of an evaluation word which is a term relating to the quality of the evaluation target and a score indicating the degree of good or bad,
The processing result configuration means includes:
Score acquisition means for acquiring a score paired with the evaluation word from the evaluation information storage means;
Using one or more scores for the same evaluation object acquired by the evaluation object acquisition means, an evaluation calculation means for calculating an evaluation for the evaluation object;
The advertisement output system according to claim 3, further comprising a score processing result configuring unit that configures a processing result using the evaluation target acquired by the evaluation target acquiring unit and the score calculated by the evaluation calculating unit. The natural language processing unit includes:
A change information storage means for storing one or more change information that is a pair of evaluation change information that is information for changing an evaluation word and a change word that indicates an evaluation change with respect to the evaluation word;
A change word acquisition means for acquiring a change word within a predetermined number of characters from the evaluation word;
Evaluation change information acquisition means for acquiring evaluation change information paired with the change word;
For the score acquired by the score acquisition means, using the evaluation change information acquired by the evaluation change information acquisition means, further changing the score, further comprising a change score calculation means for calculating a new score,
The evaluation calculation means includes
Using the one or more scores for the same evaluation object acquired by the evaluation object acquisition means and the one or more new scores for the same evaluation object calculated by the modified score calculation means, the evaluation for the evaluation object The advertisement output system according to claim 4, wherein: The sentence division control means includes:
Cutting out a third character string of a predetermined size (n characters [n is an integer of 2 or more]) from the beginning of the sentence constituting the sentence;
It is a character string constituting the cut out third character string, and it is determined whether the character string exists in the dictionary information storage means in order from the character string of n characters from the head, and the character string of n characters If the character string does not exist, it is determined whether or not a character string of (n-1) characters from the beginning exists in the dictionary information storage means, and this determination is made by reducing the number of characters one character at a time. Or if there is no character string, make a decision until the first character is reached,
When it is determined that the character string exists, the function word information, the pre-connection symbol, and the post-connection symbol corresponding to the character string are read from the dictionary information storage unit, and the read function word information, the pre-connection symbol, and the back Add a connection symbol to the string,
If the character string does not exist, an element dividing unit that assigns unknown word information that is information indicating an unknown word to the character string of one character;
The element dividing means is a continuous unknown word to which unknown word information is added, concatenates hiragana characters to form a character string, and is a continuous unknown word that concatenates characters that are not hiragana to form a character string. ,
Giving collective information that is information specifying the type of character string indicating the concatenated character string or the character of the unknown word that has not been concatenated;
Read two adjacent elements divided by the element dividing means or group information,
Of the read element or group information, the preceding element or group information and the following element or group are identified from the preceding element or group information after connection symbol and the subsequent element or group information preceding connection symbol. Determine whether the information can be linked,
Element linking means for collecting a plurality of second character strings that are the combined character strings by collecting connectable elements or group information into one character string;
From the function word information corresponding to the last first character string of each of the plurality of second character strings acquired by the element linking means or a post-connection symbol, to determine information specifying the type of each second character string,
4. From the character string acquisition means which arrange | positions the information which specifies the kind of each said several 2nd character string, and the some information which made the said each 2nd character string a pair on memory. The advertisement output system according to claim 5. The server device
A sentence storage unit storing sentences;
A sentence acquisition unit that acquires a sentence according to a processing result in the natural language processing unit from the sentence storage unit;
A server-side text transmission unit that transmits the text acquired by the text acquisition unit to the information terminal device;
The information terminal device
A terminal-side text receiver that receives text sent by the server device;
The advertisement output system according to any one of claims 1 to 6, further comprising a sentence output unit that outputs a sentence received by the terminal-side sentence receiving unit. The advertisement output system according to any one of claims 1 to 5, further comprising a second information terminal device,
The second information terminal device
A second sentence receiving unit for receiving sentences;
A second terminal-side text transmission unit that transmits the text received by the second text reception unit to the information terminal device;
A second terminal side text receiving unit for receiving text from the information terminal device;
A second sentence output unit for outputting the sentence received by the second terminal side sentence receiver;
The information terminal device
A terminal-side text receiver that receives text from the second information terminal device;
The advertisement output system according to any one of claims 1 to 6, further comprising a sentence output unit that outputs a sentence received by the terminal-side sentence receiving unit. The server device
An advertisement information storage unit storing advertisement information corresponding to the term;
The advertisement information acquisition unit
The advertisement output system according to claim 1, wherein advertisement information corresponding to a term included in a processing result acquired by the natural language processing unit is acquired from the advertisement information storage unit. The server device
An advertisement information storage unit storing advertisement information corresponding to the category;
The advertisement information acquisition unit
Category conversion dictionary storage means for storing a category conversion dictionary indicating correspondence between terms and categories;
Category acquisition means for acquiring a category corresponding to a term included in the processing result acquired by the natural language processing section from the category conversion dictionary storage section means;
The advertisement output system according to any one of claims 1 to 8, further comprising advertisement information acquisition means for acquiring advertisement information corresponding to the category acquired by the category acquisition means from the advertisement information storage unit. A dialog link information storage unit that stores dialog link information that includes first dialog information corresponding to the first dialog sentence and second dialog information corresponding to the second dialog sentence; Equipped,
The sentence storage unit
A first dialogue sentence that is a dialogue sentence that constitutes the first dialogue information, a second dialogue sentence that is a dialogue sentence that constitutes the second dialogue information, respectively, the first dialogue information, Stored in association with the second dialogue information,
The sentence acquisition unit
Using the processing result in the natural language processing unit, search the dialogue linkage information storage unit, to obtain the first dialogue information acquisition means,
The advertisement output system according to claim 7, further comprising a dialog text acquisition unit that acquires a second dialog text corresponding to the second dialog information paired with the first dialog information from the text storage unit. The dialogue linkage information is
First dialogue information, which is information composed of one or more second character strings corresponding to the first dialogue sentence or the one or more second character strings;
The advertisement output system according to claim 11, further comprising second dialogue information which is one or more second character strings corresponding to the second dialogue sentence or information composed of the one or more second character strings. The dialogue linkage information is
The advertisement output system according to claim 12, further comprising priority information corresponding to the first dialogue information and priority information corresponding to the second dialogue information. An advertisement output method realized by a sentence receiving unit, a natural language processing unit, an advertisement information acquiring unit, and an advertisement information transmitting unit,
A text receiving step for receiving text from the information terminal device;
A natural language processing step of performing natural language processing on the text received in the text receiving step and obtaining a processing result including at least one term;
Using the processing result acquired in the natural language processing step, an advertising information acquisition step for acquiring advertising information;
An advertisement output method comprising an advertisement information transmission step of transmitting the advertisement information acquired in the advertisement information acquisition step to the information terminal device. In the natural language processing step,
From the sentence, obtaining preference information that is information about the preference of the input user of the sentence,
In the advertisement information acquisition step,
The advertisement output method according to claim 14, wherein advertisement information that matches the preference information is acquired. A first character string, function word information that is information specifying the function or type of the first character string in the sentence, and a pre-connection symbol that is information specifying the type of character string that can be connected before the first character string; , A character string that can be connected after the first character string, or a symbol indicating that it is a delimiter, and a post-connection symbol that is information specifying the type of the second character string in the sentence delimited by the symbol One or more dictionary information having a pair of and are stored in a storage medium,
One or more evaluation information having evaluation words, which are terms related to the quality of the evaluation target, are stored in a storage medium,
The natural language processing step includes
The first character string corresponding to the post-connection symbol indicating that the sentence is a delimiter is extracted from the sentences constituting the sentence, and the sentence is divided into two or more second character strings that are two or more character strings using the first character string as a key. , From the function word information corresponding to the last first character string of the second character string or the post-connection symbol, the information for identifying the type of the second character string is acquired, and the second character string and the second character string A sentence division control step for obtaining information for identifying the type of
An evaluation object acquisition step of identifying a second character string including an evaluation object from the type of the second character string acquired in the sentence division control step, and acquiring an evaluation object from the second character string;
A second character string that can include an evaluation word is specified using another second character string type in a sentence having the second character string including the evaluation target, and the evaluation word that the second character string has is stored in the storage medium. An evaluation word acquisition step of searching and acquiring from
The advertisement output method according to claim 15, further comprising a processing result configuration step of configuring a processing result using the evaluation target acquired in the evaluation target acquisition step and the evaluation word acquired in the evaluation word acquisition step. Computer
A server-side text receiving unit that receives text from the information terminal device;
A natural language processing unit that performs natural language processing on the text received by the server side text receiving unit, and obtains a processing result including at least one term;
Using the processing result acquired by the natural language processing unit, an advertising information acquisition unit that acquires advertising information;
The program for functioning as an advertisement information transmission part which transmits the advertisement information which the said advertisement information acquisition part acquired to the said information terminal device. The natural language processing unit includes:
From the sentence, obtaining preference information that is information about the preference of the input user of the sentence,
The advertisement information acquisition unit
18. The program according to claim 17, which causes a computer to function so as to acquire advertisement information that matches the preference information. A first character string, function word information that is information specifying the function or type of the first character string in the sentence, and a pre-connection symbol that is information specifying the type of character string that can be connected before the first character string; , A character string that can be connected after the first character string, or a symbol indicating that it is a delimiter, and a post-connection symbol that is information specifying the type of the second character string in the sentence delimited by the symbol One or more dictionary information having a pair of and are stored in a storage medium,
One or more evaluation information having evaluation words, which are terms related to the quality of the evaluation target, are stored in a storage medium,
The natural language processing unit includes:
Evaluation information storage means for storing one or more evaluation information having evaluation words, which are terms related to the quality of the evaluation object;
The first character string corresponding to the post-connection symbol indicating that the sentence is a delimiter is extracted from the sentences constituting the sentence, and the sentence is divided into two or more second character strings that are two or more character strings using the first character string as a key. , From the function word information corresponding to the last first character string of the second character string or the post-connection symbol, the information for identifying the type of the second character string is acquired, and the second character string and the second character string Sentence division control means for acquiring information identifying the type of
An evaluation object acquisition means for specifying a second character string including an evaluation object from the type of the second character string acquired by the sentence division control means, and acquiring an evaluation object from the second character string;
A second character string that can include an evaluation word is specified using another second character string type in a sentence having the second character string that includes the evaluation target, and the evaluation word that the second character string has is the evaluation information. An evaluation word acquisition means for searching and acquiring from the storage means;
19. The computer according to claim 18, wherein the computer functions so as to include a processing result configuration unit configured to configure a processing result using the evaluation target acquired by the evaluation target acquisition unit and the evaluation word acquired by the evaluation word acquisition unit. program. The server apparatus which comprises the advertisement output system in any one of Claims 1-13.

Images