JP2012168863A - Behavior information recording apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow a user to easily perform an objective self-analysis.SOLUTION: A collection unit 21 collects at least one of character information and speech information input in a terminal device by a user through a network. A behavior information generation unit 23 generates behavior information on the user from information collected by the collection unit 21. A behavior information accumulation unit 12 accumulates the behavior information generated by the behavior information generation unit 23. An analysis unit 29 performs a character analysis from the behavior information accumulated by the behavior information accumulation unit 12.

Description

  The present invention relates to a behavior information recording apparatus that records and accumulates human behavior.

  Conventionally, attempts have been made to collect human behavior patterns and use them in marketing strategies (see, for example, Patent Document 1). As IT infrastructure is developed year by year, companies are able to make quantitative judgments based on more accurate data.

  On the other hand, the number of users who use blogs, miniblogs (Twitter (registered trademark), etc.), and SNS (Social Network Service) in a narrow sense (hereinafter simply referred to as SNS in the present specification) It is increasing. Along with this, the number of users who use blogs instead of handwritten diaries and notebooks is increasing. In SNS, an avatar is increasingly used as a character that is a substitute for itself.

JP 2002-92253 A

  Some people want to know what kind of person they are objectively. However, the subjectivity is easily mixed in self-evaluation and Q & A personality diagnosis tests that the self thinks about themselves, which may be very different from the evaluation of others.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a technique for easily performing objective self-analysis.

  A behavior information recording device according to an aspect of the present invention includes a collection unit that collects at least one of character information and voice information input to a terminal device by a user via a network, and information collected by the collection unit. A behavior information generation unit that generates behavior information, a storage unit that stores behavior information generated by the behavior information generation unit, an analysis unit that performs person analysis from the behavior information stored by the storage unit, and a virtual on a network A virtual world construction unit that constructs the world; and an avatar operation control unit that causes the user's avatar to perform an autonomous operation reflecting the user's analysis in the virtual world.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, and the like are also effective as an aspect of the present invention.

  According to the present invention, objective self-analysis can be easily performed.

It is a block diagram which shows the whole structure of the action information recording system which concerns on embodiment of this invention. It is a block diagram which shows the structural example of a portable terminal device. It is a block diagram which shows the structural example of an installation type terminal device. It is a block diagram which shows the structure of the action information recording device which concerns on Example 1. FIG. It is a figure which shows an example of the table constructed | assembled in an action information storage part. It is a flowchart for demonstrating the process until it accumulate | stores the collected information as action information in an action information storage part. It is a figure which shows the example which rendered the user's action information superimposed in the virtual world. It is a flowchart for demonstrating the process which superimposes and draws a user's action information in a virtual world. FIGS. 9A to 9C are diagrams illustrating an example of a blog generated by the weblog content generation unit. It is a block diagram which shows the structure of the action information recording device which concerns on Example 2. FIG. It is a figure which shows an example of a person analysis chart. 12 is a table that defines keywords and key phrases that are the basis of parameter adjustment in FIG. 11. It is a figure for demonstrating the specific example of the writing site regarding a person. It is a figure which shows an example of the action pattern table which described the action pattern of a day-of-week unit. It is a figure which shows an example of a specific day, a ToDo list | wrist, and an achievement rate. It is a figure which shows an example of the message which prompts execution of the action which should be taken by the message production | generation part. FIGS. 17A to 17F are diagrams for explaining an example of a data structure constructed in the operation data holding unit. It is a flowchart for demonstrating the basic process of the avatar's autonomous operation | movement by an avatar operation control part. It is a flowchart for demonstrating the basic process of the autonomous operation | movement of an avatar by the avatar operation control part at the time of receiving an action command. FIG. 10 is a block diagram for explaining a modification of the second embodiment. It is a flowchart for demonstrating the conversation process of the user of a real world, and the avatar of an absent user.

  FIG. 1 is a block diagram showing an overall configuration of a behavior information recording system 500 according to an embodiment of the present invention. The behavior information recording system 500 according to the present embodiment includes the behavior information recording device 100 and a plurality of terminal devices. The behavior information recording device 100 and each of the plurality of terminal devices can perform bidirectional communication via a network (hereinafter, assumed to be the Internet) 300.

  The behavior information recording device 100 is constructed by at least one server. In FIG. 1, terminal devices are drawn by being classified into a stationary terminal device 201 and a mobile terminal device 202. Moreover, the terminal device which three users (user AC) hold | maintain among many users is drawn. The installed terminal devices 201A to 201C are terminal devices installed in the users A to C, respectively. Examples of the installation type terminal device include a PC, a TV, and a game machine. In the following description, a PC is assumed. Users A to C carry mobile terminal devices 202A to 202C, respectively. Examples of the portable terminal device include a mobile phone, a smartphone, and a portable game machine. In the following description, a smartphone is assumed.

  The stationary terminal device 201 can be connected to the network 300 via a wired line (for example, an optical fiber line, an ADSL line, a CATV line) or a wireless line (for example, a wireless LAN). The mobile terminal device 202 can be connected to the network 300 via a wireless line (for example, a wireless LAN or a mobile phone network). The stationary terminal device 201 and the portable terminal device 202 can exchange data via a cable (for example, a USB cable), short-range wireless communication (for example, Bluetooth (registered trademark), a memory card, or the network 300).

  FIG. 2 is a block diagram illustrating a configuration example of the mobile terminal device 202. The portable terminal device 202 includes an antenna 41, a wireless communication unit 42, a position detection unit 43, a photographing unit 44, an external input / output unit 45, a character input unit 46, a voice input unit 47, a display unit 48, a voice output unit 49, a clock. Unit 50, processing unit 51, and storage unit 52.

  The wireless communication unit 42 transmits and receives radio waves in a frequency band corresponding to the wireless communication method via the antenna 41. Although one antenna 41 is drawn in FIG. 2 for the sake of simplicity of explanation, a plurality of antennas 41 are actually provided according to the type of the wireless communication method. For example, an antenna for communicating with a base station via a mobile phone network, an antenna for communicating with an access point via a wireless LAN, and an antenna for short-range wireless communication with a device located in a short distance are provided separately. It is done. The wireless communication unit 42 modulates a signal input from the processing unit 51 and transmits a radio wave from the antenna 41. In addition, the signal received by the antenna 41 is demodulated and output to the processing unit 51.

  The position detection unit 43 detects or estimates the position of the mobile terminal device 202 and outputs the position information to the processing unit 51. The position detection unit 43 includes a GPS (Global Positioning System), for example, and calculates latitude and longitude based on a plurality of signals received from GPS satellites. Note that the current position of the mobile terminal device 202 can be estimated based on communication with a base station or an access point in addition to using GPS.

  The imaging unit 44 captures a still image or a moving image due to a user action. The imaging unit 44 outputs the captured image to the processing unit 51 as image information. At this time, it is preferable that position information detected by the position detection unit 43 is added to the image information. Note that recording by the audio input unit 47 may be performed simultaneously with moving image shooting by the shooting unit 44.

  The external input / output unit 45 is an interface for connecting a cable to an external device (for example, the installation type terminal device 201). Not only data exchange but also power supply can be received through the cable.

  The character input unit 46 receives character input by a user action. The character input unit 46 may be a touch panel input method, a physical key input method, or both. The character input unit 46 outputs the received character information to the processing unit 51. At this time, it is preferable that position information detected by the position detection unit 43 is added to the character information.

  The voice input unit 47 receives an external sound. In the present embodiment, voice input mainly by a user's speech during a call or a user's speech during the recording function is accepted. The voice input unit 47 outputs the received voice information to the processing unit 51. At that time, it is preferable that position information detected by the position detector 43 is added to the audio information.

  The display unit 48 displays the image generated or reproduced by the processing unit 51. The audio output unit 49 displays the image generated or reproduced by the processing unit 51. The clock unit 50 outputs date / time information to the processing unit 51.

  The processing unit 51 comprehensively controls the entire mobile terminal device 202. The storage unit 52 holds various programs and data. In the present embodiment, the processing unit 51 mainly uses the clock unit 50 for position information input from the position detection unit 43, character information input from the character input unit 46, and voice information input from the voice input unit 47. The date and time information input from is added and temporarily stored in the storage unit 52.

  The processing unit 51 transmits the information temporarily stored in the storage unit 52 to the behavior information recording device 100 via the network 300 together with the identification information of the mobile terminal device 202. At that time, it may be transmitted by directly connecting to the network 300 using the wireless communication unit 42 and the antenna 41, or such information is once transmitted to the stationary terminal device 201, via the stationary terminal device 201. Then, it may be transmitted by connecting to the network 300.

  In the former case, the information may be sent every time information is acquired, may be sent when a certain amount of information is accumulated, or may be sent periodically (for example, every other hour). Good. In the latter case, the information acquired in the storage unit 52 may be sent in a batch when the portable terminal device 202 and the stationary terminal device 201 are connected to a cable or the like.

  FIG. 3 is a block diagram illustrating a configuration example of the stationary terminal device 201. The installed terminal device 201 includes a communication unit 42a, an external input / output unit 45, a character input unit 46, a voice input unit 47, a display unit 48, a voice output unit 49, a clock unit 50, a processing unit 51, and a storage unit 52. . These components are basically the same as the corresponding components of the mobile terminal device 202 shown in FIG. The specifications of these components are higher than the specifications of the corresponding components of the mobile terminal device 202 shown in FIG. 2, and are excellent in computing ability and user interface.

  FIG. 4 is a block diagram illustrating the configuration of the behavior information recording apparatus 100 according to the first embodiment. The behavior information recording apparatus 100 includes a communication unit 10, an administrator console 11, a behavior information storage unit 12, a map data storage unit 13, a 3D object storage unit 14, a web log data storage unit 15, and a processing unit 20.

  The processing unit 20 includes a collection unit 21, a phonetic character conversion unit 22, a behavior information generation unit 23, a site management unit 24, a virtual world construction unit 25, a behavior information drawing unit 26, a behavior reproduction unit 27, and a weblog content generation unit 28. Including. These configurations can be realized by an arbitrary processor, memory, or other LSI in terms of hardware, and can be realized by a program loaded in the memory in terms of software, but here by their cooperation. Draw functional blocks. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.

  The communication unit 10 is connected to the network 300 via a wired line. The administrator console 11 is an administrator interface for the administrator of the behavior information recording device 100 to operate the behavior information recording device 100. It is used for maintenance management of the behavior information recording apparatus 100.

  The behavior information accumulation unit 12 accumulates the behavior information generated by the behavior information generation unit 23. The map data holding unit 13 holds map data. The map data may be two-dimensional data or three-dimensional data. In the following description, an example using three-dimensional data will be described. It is assumed that the map data includes latitude / longitude information of each point on the map. The 3D object holding unit 14 holds three-dimensional object data to be drawn in a three-dimensional space described by three-dimensional map data. For example, each user's avatar data is held. In addition, since the avatar of each user can be created by an existing service, the description thereof is omitted in this specification. The web log data storage unit 15 stores blog data (including mini-blog data) of each user.

  The processing unit 20 comprehensively controls the entire behavior information recording apparatus 100. The site management unit 24 operates a social networking site on the network 300. In the first embodiment, an avatar creation service and a blog service are also provided to the user. The virtual world construction unit 25 constructs a virtual world corresponding to the real world on the network 300 in association with the social networking site operated by the site management unit 24. More specifically, a virtual world imitating an actual city is constructed using the map data held in the map data holding unit 13.

  The collection unit 21 collects at least one of character information and voice information input to the mobile terminal device 202 by the user and position information of the mobile terminal device 202 via the network 300. The collection unit 21 collects image information captured by the user using the mobile terminal device 202 via the network 300. This image information may include audio information. Note that the collected date information, voice information, and image information are basically appended with date and time information and position information at the time when they are input to the processing unit 51 of the mobile terminal device 202.

  The voice character conversion unit 22 converts the voice information collected by the collection unit 21 into character information. This can be performed by using a general voice recognition technique. This phonetic character conversion may be executed on the mobile terminal device 202 or the stationary terminal device 201 side. In that case, the action information recording apparatus 100 does not require the phonetic character conversion unit 22.

  The behavior information generation unit 23 generates user behavior information from the character information, voice information, image information, position information, and time information collected by the collection unit 21. For example, the collected character information and the character information converted by the phonetic character conversion unit 22 are subjected to text mining to extract a sentence including a keyword or key phrase related to “5W1H” useful for identifying the user's action. In Japanese, morphological analysis is performed prior to text mining. Morphological analysis is a process that divides a sentence into meaningful minimum unit character strings and classifies the divided character strings into parts of speech. In the present specification, the details of the text mining itself are not focused on, and thus detailed description thereof is omitted.

  The behavior information generation unit 23 also considers information collected from the mobile terminal devices 202 of other users registered in the social networking site, which is presumed to have contact with the target user, and the behavior information of the target user. Can also be generated. For example, when the destination of the call that is the basis of the voice information of the target user or the transmission destination of the e-mail that is the basis of the character information is the other user, the behavior information generation unit 23 is the mobile terminal of the other user. Of the information collected from the device 202, information related to the behavior of the target user can be used. More specifically, the text including the name of the target user is extracted from the character information collected from the mobile terminal device 202 of the other user (including information subjected to the phonetic character conversion).

  The behavior information generation unit 23 stores the generated behavior information in the behavior information storage unit 12 for each user. FIG. 5 is a diagram illustrating an example of a table constructed in the behavior information storage unit 12. In the table shown in FIG. 5, “user”, “date / time”, “position information”, “character information”, “voice information”, and “image information” are defined as items. The behavior information generation unit 23 embeds information in each item cell for each collected information. Note that the acquisition method and data classification are described for the cells of “character information”, “voice information”, and “image information”. The actual data of “text information”, “voice information” and “image information” is recorded in a separate recording area, and the cells of “character information”, “voice information” and “image information” are stored in the corresponding recording area. Associated with an address.

  FIG. 6 is a flowchart for explaining processing until the collected information is accumulated as behavior information in the behavior information accumulation unit 12. The collection unit 21 acquires character information or voice information from the mobile terminal device 202 (S10). Note that when the collected information is single position information not accompanying character information or the like and single image information not accompanied by voice information, the information is directly stored in the behavior information storage unit 12.

  The phonetic character conversion unit 22 determines whether the information collected by the collecting unit 21 is voice information (S11). If it is voice information (Y in S11), the voice information is converted into character information (S12). When it is not voice information (N of S11), step S12 is skipped.

  The behavior information generation unit 23 determines whether or not the number of characters included in the character information exceeds the set number of characters (S13). If the set number of characters is exceeded (Y in S13), the sentence is summarized (S14). The behavior information generation unit 23 summarizes sentences using general sentence summarization software. If the number of set characters is not exceeded (N in S13), step S14 is skipped. The behavior information generation unit 23 stores character information in the behavior information storage unit 12 together with the date information and the position information (S15).

  Returning to FIG. The behavior information drawing unit 26 superimposes the user behavior information stored in the behavior information storage unit 12 at a corresponding place in the virtual world. The behavior reproducing unit 27 causes the user's avatar to perform an operation according to the behavior information of the user accumulated in the behavior information accumulating unit 12 in the virtual world. These drawings and playback using avatars are basically executed on a daily basis, but may be executed on a time span of less than one day or more than one day.

  FIG. 7 is a diagram illustrating an example in which user behavior information is drawn in a superimposed manner in the virtual world. The behavior information shown in FIG. 7 is a part of the behavior information on July 25, 2010. At 9:00 of the day, the user writes “buy a cigarette and go to work” on the miniblog using the mobile terminal device 202 at the point P1. Thereafter, at 9:10, the user is calling someone using the mobile terminal device 202 at the point P2. The content of the conversation includes the remark "In front of the Sony building in Ginza at 20:00." After that, at 9:20, a moving image is shot using the camera function of the mobile terminal device 202 at a point P3. The subject of the movie is a jellyfish. The user also said, “Oh, the jellyfish on the canal are beautiful.” In addition, the action reproducing unit 27 can move the avatar as time elapses. That is, the action of the day can be played back as a moving image. The record of action information based on these maps can be browsed by the person himself or a person authorized to view it.

  FIG. 8 is a flowchart for explaining a process of drawing user behavior information in a virtual world in a superimposed manner. The behavior information generation unit 23 acquires map data from the map data holding unit 13 (S20), and the virtual world construction unit 25 constructs a virtual world based on the map data. The behavior information generation unit 23 acquires data for drawing an avatar from the 3D object holding unit 14 (S21). The behavior information generation unit 23 acquires the daily behavior information of the target user from the behavior information storage unit 12 in order of time (S22). The behavior information generation unit 23 places an avatar at a position specified by the position information included in the acquired behavior information (S23), and arranges character information and / or image information included in the acquired behavior information in the vicinity thereof. (S24). The behavior information generation unit 23 determines whether or not all the behavior information of the target user for one day has been acquired (S25). When not acquiring (N of S25), it changes to step S22 and repeats the process of steps S23-S25. When acquired (Y of S25), the drawing process of the action information for one day of the target user is ended.

  Returning to FIG. The web log content generation unit 28 generates at least one of a sentence, an image, and a sound to be posted on the web log from the behavior information generated by the behavior information generation unit 23, and accumulates it in the web log data accumulation unit 15. . Here, it is not based on a map as shown in FIGS. 7 and 8, but is an automated work that a user has manually performed on an existing platform.

  FIGS. 9A to 9C are diagrams illustrating an example of a blog generated by the weblog content generation unit 28. FIG. FIG. 9A shows a real pattern blog, FIG. 9B shows a serious pattern blog, and FIG. 9C shows a fuzzy pattern blog. The user can select one of these patterns.

  As shown in FIG. 9A, a blog having a real pattern is created by using character information and voice information input from the mobile terminal device 202 by the user, which are stored in the behavior information storage unit 12, almost as they are. It is.

  As shown in FIG. 9 (b), in addition to the text described in the real pattern blog, the serious pattern blog is the user's opinions and insights on the theme of the text, knowledge obtained from the knowledge database site on the web, Or a scenario obtained from a scenario database site on the web is added. The web log content generation unit 28 detects a keyword of each sentence, extracts other sentences of the user related to the keyword, extracts knowledge about the keyword from a knowledge database site on the web, Such a blog can be generated by extracting a scenario related to the keyword from a scenario database site on the web.

  As shown in FIG. 9C, the blog of the fuzzy pattern is added with the user's impression on the sentence theme or a scenario obtained from a scenario database site on the web. The web log content generation unit 28 detects a keyword of each sentence, extracts other sentences of the user related to the keyword, or extracts a scenario related to the keyword from a scenario database site on the web. Can generate such a blog. Further, the web log content generation unit 28 can convert or process the style using style conversion software.

  As described above, according to the behavior information recording apparatus 100 according to the first embodiment, by accumulating character information, voice information, or image information input by the user through the mobile terminal device 202, the user's behavior can be accurately determined. Easy to record. Further, if date / time information and position information when such information is input are also accumulated, the action can be recorded more accurately. Moreover, the accuracy can be further improved by comparing with the action information of the registered user of the same social networking site.

  Also, by displaying the action information in a three-dimensional virtual world corresponding to the real world, the action record can be confirmed more intuitively. Moreover, an action record can be confirmed with a moving image by making an avatar act along own action record. Also, by automatically generating a blog based on the accumulated behavior information, it is possible to save the trouble of creating a blog.

  FIG. 10 is a block diagram illustrating the configuration of the behavior information recording apparatus 100 according to the second embodiment. The behavior information recording apparatus 100 includes a communication unit 10, an administrator console 11, a behavior information storage unit 12, a map data storage unit 13, a 3D object storage unit 14, a motion data storage unit 16, a person analysis storage unit 17, and a behavior pattern storage. Unit 18, ToDo list holding unit 19, and processing unit 20.

  The processing unit 20 includes a collection unit 21, a phonetic character conversion unit 22, a behavior information generation unit 23, a site management unit 24, a virtual world construction unit 25, an analysis unit 29, an avatar operation control unit 30, a reception unit 31, and a behavior history generation unit. 32, an achievement rate calculation unit 33 and a message generation unit 34 are included. These configurations can be realized by an arbitrary processor, memory, or other LSI in terms of hardware, and can be realized by a program loaded in the memory in terms of software, but here by their cooperation. Draw functional blocks. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.

  Description of components common to the components in FIG. 10 and the components in FIG. 4 will be omitted as appropriate. The motion data holding unit 16 holds motion data for operating the avatar in the virtual world. The person analysis storage unit 17 stores the person analysis information of each user analyzed by the analysis unit 29. The behavior pattern accumulation unit 18 accumulates each user's behavior pattern specified by the analysis unit 29. The ToDo list holding unit 19 holds a list in which actions to be performed on a specific date of each user are described.

  The analysis unit 29 performs data analysis by mining the behavior information accumulated by the behavior information accumulation unit 12. For example, as described above, the character information in the accumulated behavior information is subjected to morphological analysis, classified as part of speech, and then text mined to extract a keyword or key phrase. The analysis unit 29 performs person analysis of each user based on the keyword or key phrase.

  FIG. 11 shows an example of a person analysis chart. This chart analyzes a person with four parameters. The first parameter is a parameter related to vitality, and indicates whether it is an outward type or an inward type. The second parameter is a parameter related to information collection, and indicates whether it is a real type or an intuitive type. The third parameter is a parameter related to decision, and indicates whether it is a thinking type or an emotional type. The fourth parameter is a parameter related to the lifestyle, and indicates whether it is a normative type or a flexible type. In FIG. 11, each parameter is classified into nine stages. The analysis unit 29 adjusts the value of each parameter.

  FIG. 12 is a table that defines keywords and key phrases that are the basis of parameter adjustment in FIG. For example, the analysis unit 29 counts keywords “speak”, “speak”, “associate”, “go out”,... Which are keywords that cause the user to be inferred from the character information in the behavior information of the target user. Then, the first parameter is moved from the current value (initial value is “1”) in the addition direction according to the count number. Similarly, the analysis unit 29 selects “listening”, “thinking”, “concentration”, “one person”,..., Which are keywords for inferring the introverted type from the character information in the behavior information of the target user. Counting is performed, and the first parameter is moved from the current value in the subtraction direction in accordance with the count number. These processes are also performed for other parameters.

  The number of keywords required to move a parameter by one unit and the period range of data to be analyzed can be arbitrarily set by a designer. Further, weighting may be performed to reflect the latest tendency more strongly. For example, a keyword detected from character information close to the present may be handled as a larger value (greater than 1).

  Returning to FIG. The site management unit 24 operates a writing site related to a person on the network 300. The analysis unit 29 performs person analysis of the user in consideration of writing for the user written in the writing site. That is, a keyword or key phrase is extracted from the writing.

  FIG. 13 is a diagram for explaining a specific example of a writing site related to a person. The site management unit 24 includes an inference engine 241. Members in the social networking site can register personal evaluations of family members, friends, colleagues, etc. in the writing site. At that time, in principle, the contributor posts anonymously. Since the evaluation target person needs to be accurately specified, it is preferable that as much information as possible to identify the evaluation target person is described.

  In FIG. 13, posts A1 to A3 are filled with the quantitative evaluation and qualitative evaluation by Ichiro Nomura. In order to simplify the explanation in FIG. 13, three posts are not drawn. However, as the number of posts increases, it can be expected that a more objective and accurate evaluation from others is obtained. The inference engine 241 creates an evaluation report B1 by assembling a plurality of posted postings or averaging quantitative evaluations. This evaluation report B can be browsed by the person himself or a person authorized to view the report.

  In addition, it may be a mechanism that allows a user who wants to receive a person evaluation from another person to post a person evaluation for the user when the user desires it. In addition, a privilege may be given to a user who has made a post that approximates the evaluation report generated by the inference engine 241. For example, a privilege may be given to a user who has made a posting in which the difference between the quantitative evaluation score of the evaluation report and the quantitative evaluation score in the post is less than the set value. Such benefits include points and items that can be used within the social networking site, or partial reduction or exemption of paid services.

  Returning to FIG. Based on the behavior information accumulated in the behavior information accumulation unit 12, the analysis unit 29 identifies the behavior pattern of the user's day of the week or date unit and accumulates it in the behavior pattern accumulation unit 18. FIG. 14 is a diagram illustrating an example of an action pattern table describing action patterns in units of days of the week. The analysis unit 29 extracts keywords, key phrases, and key sentences related to actions that frequently appear on a day-of-week basis by text mining the character information included in the action information stored in the action information storage unit 12. The analysis unit 29 employs keywords, key phrases, and key sentences exceeding the set number of times among the extracted keywords, key phrases, and key sentences as action patterns for each day of the week. FIG. 14 is an action pattern table generated in this way.

  Returning to FIG. Based on the user behavior information accumulated in the behavior information accumulation unit 12, the behavior history generation unit 32 aggregates the behavior of the user on a specific day and generates a behavior history of the specific day. The achievement rate calculation unit 33 compares the behavior history generated by the behavior history generation unit 32 with the ToDo list describing the behavior to be performed on a specific day, and calculates the achievement rate of the behavior to be performed on the specific day. . When the achievement rate calculated by the achievement rate calculation unit 33 is less than the set value, the message generation unit 34 generates a message that prompts execution of an action to be taken.

  FIG. 15 is a diagram illustrating an example of a specific date, a ToDo list, and an achievement rate. Registration of actions to be performed in the ToDo list is basically performed by the user. It should be noted that behaviors that should be customary, such as new year's greetings and New Year's card postings, may be set in the ToDo list by default. The user can delete or change the setting. You can also shift the date. The achievement rate calculation unit 33 calculates the achievement rate of the action to be taken from the action history of the specific day. For example, if an action has to be taken only once in the past four years, the achievement rate is 25%. The table shown in FIG. 15 can be referred to by the user at any time.

  FIG. 16 is a diagram illustrating an example of a message that is generated by the message generation unit 34 and that prompts execution of an action to be performed. This message is generated when the achievement rate of the action to be taken is less than a set value (for example, 50%). The message is notified to the user before the date of the action to be taken (for example, the previous day, three days ago, or one week ago). Examples of notification means include e-mail and posting on a blog top screen.

  Returning to FIG. The avatar operation control unit 30 causes the user's avatar to perform an autonomous operation reflecting the user's analysis in the virtual world. The avatar operation control unit 30 can autonomously operate the avatar using existing artificial intelligence technology. There are various methods of artificial intelligence technology such as neural networks and genetic algorithms, but a relatively simple model is described below.

  FIGS. 17A to 17F are diagrams for explaining an example of a data structure constructed in the operation data holding unit 16. In this example, actions to be performed by the avatar are hierarchically defined. The table shown in FIG. 17A divides the “action” of the avatar into “going out” and “at home”. The tables shown in FIGS. 17B to 17D select the destination, the moving means, the accompanying person,... When “go out” is selected in the table of FIG. It is a table to do. The table shown in FIG. 17E is a table for selecting the genre of the movie when “Movie Theater” is selected in the table shown in FIG. The table shown in FIG. 17F is a table for selecting being a house when “I am at home” is selected in the table shown in FIG.

  FIG. 18 is a flowchart for explaining basic processing of an avatar autonomous operation by the avatar operation control unit 30. First, the avatar motion control unit 30 weights the probability of random number generation with the person analysis parameter (S30). In principle, random numbers are generated with equal probability within the range of the number of choices. In the table shown in FIG. 17A, the number of options is “2”, and “1” or “2” occurs with equal probability. In step S30, the occurrence probability is adjusted. In the case of the user having the person evaluation parameter shown in FIG. 14, the first parameter relating to vitality is 1.4 (that is, slightly introverted), so the occurrence probability of “at home” is 1.4, “ The occurrence probability of “going out” is adjusted to “0.6”.

  The avatar movement control unit 30 generates a random number after adjusting the occurrence probability (S31). And the operation | movement of the avatar according to the random number is determined (S32). By repeating this basic process, the specific operation of the avatar is determined.

  Returning to FIG. The accepting unit 31 accepts a command (hereinafter referred to as a behavior command) for instructing an action to be executed by the user's avatar from the user's stationary terminal device 201 or the mobile terminal device 202 via the network 300. The avatar operation control unit 30 operates the user's avatar in the virtual world according to the program reflecting the user's person analysis and the action command received by the receiving unit 31.

  FIG. 19 is a flowchart for explaining basic processing of an avatar autonomous operation by the avatar operation control unit 30 when an action command is received. The receiving unit 31 receives an action command from the user's stationary terminal device 201 or the mobile terminal device 202 via the network 300 (S40). The avatar movement control unit 30 searches whether or not the action related to the action command exists in the movement data holding unit 16 (S41).

  If it does not exist (N in S41), a message indicating that the action cannot be executed is returned to the stationary terminal device 201 or the portable terminal device 202, and the receiving unit 31 waits for another action command (S40). When it exists (Y of S41), the avatar action control part 30 determines the action according to the received action command (42). For example, if the action command is “go to movie theater”, “movie theater” is selected in the table shown in FIG. Hereinafter, processing similar to Step S30 to Step S32 in FIG. 18 is performed for lower layers (S43 to S45).

  Note that the avatar motion control unit 30 may determine the motion of the avatar using an action pattern table as shown in FIG. 14 instead of motion control using such a random number. In the example of FIG. 14, by specifying the day of the week, the action to be executed by the avatar can be determined.

  As described above, according to the behavior information recording apparatus 100 according to the second embodiment, objective self-analysis can be easily performed by analyzing the self based on the accumulated self-behavior information. Further, as the amount of accumulated behavior information increases, more accurate analysis can be performed. Also, by browsing the evaluations written by other people on the posting site, it is possible to easily obtain evaluations from other people. In addition, writing from other people can be used as basic data for the self-analysis, and in this case, a hybrid evaluation in which analysis by a computer and analysis by a human being are combined can be obtained.

  In addition, by autonomously operating an avatar that reflects his / her personal analysis, his / her character can be observed from an objective standpoint and can be enjoyed as entertainment. It can also lead to self-reform.

  FIG. 20 is a block diagram for explaining a modification of the second embodiment. In the modification, the avatar operation control unit 30 includes a conversation bot unit 310. The conversation bot unit 310 includes a conversation sentence reception unit 311, a conversation sentence analysis unit 312, and a conversation sentence response unit 313. The operation data holding unit 16 includes a conversation data holding unit 161. The conversation data holding unit 161 holds various conversation data. The conversation data is preferably updated as appropriate.

  In the modification, the avatar operation control unit 30 causes the user's avatar to perform an autonomous operation reflecting the user's analysis in the virtual world after the absence of the user. Here, the absence of the user includes a permanent absence such as death and disappearance, and a relatively long absence such as assignment to a remote place, hospitalization (a symptom of an inability to talk), imprisonment, and the like.

  The avatar operation control unit 30 realizes communication between the user's avatar and another user in the real world in accordance with a program reflecting the person analysis of the absent user. That is, communication with an absent user is realized in a pseudo manner. The communication mainly assumes conversation, but also includes facial expression changes and gestures.

  The avatar movement control unit 30 extracts the user's moustache or tone from the behavior information before the user's absence (when dying, the lifetime) accumulated in the behavior information accumulation unit 12 and reflects it in the conversation in communication. For example, the morphological analysis is performed on the character information included in the action information, and the most frequent suffix added to the end of the sentence is extracted. Many of its suffixes come from dialects.

  The conversation bot unit 310 realizes a conversation between the absent user's family and friends and the absent user's avatar in a pseudo manner. For conversational bots, methods for understanding and responding to the meaning of the other party's conversation using natural language processing techniques have been studied, but in the following, a relatively simple model will be described.

  The conversation sentence receiving unit 311 receives a conversation sentence from the user's stationary terminal device 201 or the portable terminal device 202 who wants to talk with the avatar of the absent user via the network 300. When the received information is voice information, it is converted into character information. The conversation sentence analysis unit 312 extracts one or more keywords or key phrases from the accepted conversation sentence. The conversation sentence analysis unit 312 extracts the conversation sentence that is most strongly associated with the keyword or key phrase from the conversation data holding unit 161. For example, when two keywords “grandchild” and “born” are extracted, a conversation sentence “really good” is extracted. Since the existing technology is used for associating these keywords with conversational sentences, detailed description thereof is omitted. The conversation sentence analysis unit 312 may extract the conversation sentence and then modify the conversation sentence using the tone, or add the mustache to the conversation sentence. The conversation sentence analysis unit 312 returns the conversation sentence as character information or voice information to the user's stationary terminal device 201 or the portable terminal device 202.

  FIG. 21 is a flowchart for explaining a conversation process between a real-world user and an absent user's avatar. The conversation sentence receiving unit 311 receives a conversation sentence input to the stationary terminal device 201 or the portable terminal device 202 by a user who wants to talk to the absent user's avatar (S50). The conversation sentence analysis unit 312 detects one or more keywords or key phrases from the accepted conversation sentence (S51). The conversation sentence analysis unit 312 extracts the conversation sentence having the highest matching degree with the keyword or key phrase from the conversation data holding unit 161 (S52). The conversation sentence analysis unit 312 returns the extracted conversation sentence to the user's stationary terminal device 201 or the portable terminal device 202 (S53). The conversation bot unit 310 determines whether or not the conversation has ended (S54). If the conversation has not ended (N in S54), the process proceeds to step S50, and the processes in steps S50 to S54 are repeatedly executed. When it is finished (Y in S54), the conversation process is finished.

  Returning to FIG. The analysis unit 29 identifies a behavior pattern before the user's absence based on the behavior information before the user's absence accumulated in the behavior information accumulation unit 12. For example, an action pattern table as shown in FIG. 14 is generated. The avatar operation control unit 30 causes the avatar of the user to perform an autonomous operation that reflects the behavior pattern before the user is absent.

  As described above, according to the behavior information recording apparatus 100 according to the modified example of the second embodiment, by leaving an avatar reflecting its personality in the virtual world on the network after the absence, You can leave a strong impression of yourself when you exist. In addition, family members and friends can communicate with absentees in a pseudo manner.

  The present invention has been described based on some embodiments. It is understood by those skilled in the art that these embodiments are exemplifications, and that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are also within the scope of the present invention. By the way.

  In the embodiment described above, the analysis unit 29 has mainly described a technique for performing character analysis by text mining character information included in the behavior information accumulated in the behavior information accumulation unit 12. In addition to or instead of this, person analysis may be performed using psychological analysis obtained from the subject of the image information stored in the behavior information storage unit 12.

  DESCRIPTION OF SYMBOLS 100 Action information recording device, 201 Installation type terminal device, 202 Mobile terminal device, 201A Installation type terminal device, 202A Mobile terminal device, 201B Installation type terminal device, 202B Mobile terminal device, 201C Installation type terminal device, 202C Mobile terminal device, DESCRIPTION OF SYMBOLS 10 Communication part, 11 Administrator console, 12 Action information storage part, 13 Map data holding part, 14 3D object holding part, 15 Weblog data storage part, 16 Action data holding part, 161 Conversation data holding part, 17 Person analysis storage Unit, 18 behavior pattern storage unit, 19 ToDo list holding unit, 20 processing unit, 21 collection unit, 22 phonetic character conversion unit, 23 behavior information generation unit, 24 site management unit, 241 inference engine, 25 virtual world construction unit, 26 Behavior information drawing section 27 behavior reproducing unit, 28 web log content generating unit, 29 analyzing unit, 30 avatar operation control unit, 310 conversation bot unit, 311 conversation sentence receiving unit, 312 conversation sentence analyzing unit, 313 conversation sentence response unit, 31 receiving unit, 32 Action history generation unit, 33 achievement rate calculation unit, 34 message generation unit, 41 antenna, 42 wireless communication unit, 42a communication unit, 43 position detection unit, 44 shooting unit, 45 external input / output unit, 46 character input unit, 47 voice Input unit, 48 display unit, 49 audio output unit, 50 clock unit, 51 processing unit, 52 storage unit, 300 network, 500 action information recording system.

Claims (6)

A collection unit that collects at least one of character information and voice information input to a terminal device by a user via a network;
From the information collected by the collection unit, a behavior information generation unit that generates the user behavior information,
An accumulation unit that accumulates the behavior information generated by the behavior information generation unit;
From the behavior information accumulated by the accumulation unit, an analysis unit that performs person analysis,
A virtual world building department that builds a virtual world on the network;
An avatar operation control unit for causing the user's avatar to perform an autonomous operation reflecting the user's analysis in the virtual world;
A behavior information recording apparatus comprising: A site management department that operates a writing site about people on the network,
The behavior information recording apparatus according to claim 1, wherein the analysis unit analyzes a person of the user in consideration of writing to the user written in the writing site. A reception unit that receives a command for instructing an action to be executed by the user's avatar from the user's terminal device;
The avatar movement control unit operates the user's avatar in the virtual world according to a program reflecting a user analysis of the user and a command received by the reception unit. 2. The behavior information recording device according to 2.   The behavior information recording apparatus according to claim 1, wherein the virtual world construction unit constructs a virtual world corresponding to the real world on a network. Based on the user behavior information stored in the storage unit, an action history generation unit that aggregates the behavior of the user on a specific day and generates an action history on the specific day;
An achievement rate calculating unit that calculates an achievement rate of an action to be performed on the specific day by comparing the action history with a list in which an action to be performed on the specific day is described. Item 5. The behavior information recording device according to any one of Items 1 to 4.   The message generation unit according to claim 5, further comprising: a message generation unit that generates a message that prompts execution of the action to be performed when the achievement rate calculated by the achievement rate calculation unit is less than a set value. Behavior information recording device.