JP2010049633A - Information processor and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information processor, capable of providing a communication service safe and fun for a user based on the user's taste. <P>SOLUTION: A service providing server 10 connected to a plurality of user terminals 20 through a network 500 to provide a communication service between users includes an input information receiving part 111 which receives data of a sentence input by a user through the user terminal device 20 through the network 500; and a user taste analysis part 112 which analyzes the sentence to analyze whether a noun in the sentence is used as affirmative meaning or negative meaning, and registers the result to a user taste information storage part 122. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an information processing apparatus and a program.

  Social networking service (SNS) is a community-type membership service that promotes and supports connections between people. In SNS, there are services such as a function that allows users to share their profile and photos to members, a diary function that allows only members and friends to limit the scope of disclosure, and a community function that allows them to interact on the bulletin board by determining themes such as hobbies and regions It can be used on the Internet.

  However, although registered as a member, there are cases where some users do not update the diary very much or even access the site. Also, there was no effective way to find friends who seemed to feel good from among the members and to exchange emails.

  In Patent Document 1, an agent acting on behalf of a user on a website on the Internet used by those who have the same interest and purpose uses the knowledge information about the user's preferences to make compatibility with other users. It is disclosed that an evaluation is made. Thus, the user can search for a user who has a preference, or can find another person who has a preference without directly interacting with the other person.

JP 2002-229796 A

  However, in the method disclosed in Patent Document 1, the range of information used to analyze user preferences is limited to some information such as age, hobbies, sex, and occupation, and the user's diary. The user's recent interests and preferences that appear in books and e-mail content cannot be analyzed.

  Accordingly, an object of the present invention is to provide an information processing apparatus that can provide a communication service that is safe and secure for the user based on the user's preference.

  An information processing apparatus according to the present invention is an information processing apparatus that is connected to a terminal device of a plurality of users via a communication network and provides a communication service between the users, and the user passes through the terminal device. An input information receiving unit that receives input sentence data via the communication network, and the sentence is analyzed, and a noun in the sentence is used in a positive or negative meaning A user preference analysis unit that analyzes whether the user preference information is stored in the user preference information storage unit.

In this way, based on the analysis results such as writing to the user's diary and bulletin board, e-mail content, content returned to the question from pets, what nouns the user uses positively, Since it is possible to provide services using the user preference information storage unit that accumulates information on what nouns are used in a negative manner, it is possible to provide services that are safe and comfortable for users based on user preferences. Can be provided.
In addition, since the contents of the user preference information storage unit are constantly updated each time a user writes, what the user has recently been interested in and preferences can be used in a timely manner.

Further, the user preference analysis unit discriminates a noun included in the sentence by performing a morphological analysis of the sentence, and whether each noun is used in a positive sense in the sentence is negative. It is desirable to analyze whether the meaning is used.
Thereby, a sentence can be analyzed easily and reliably.

In addition, the user preference analysis unit analyzes the meaning of a morpheme that is in a modified or connected relationship with a noun in the sentence, thereby denying whether the noun is used in a positive sense in the sentence. It is desirable to analyze whether it is used in a typical sense.
Thereby, it is possible to avoid an incorrect interpretation and perform an accurate analysis.

In addition, it is preferable that the user preference analysis unit performs the morphological analysis after performing a process of replacing at least characters, pictograms, and approximate characters included in the sentence with normal characters.
Thereby, the precision of morphological analysis can be improved.

Furthermore, it is preferable that the user preference analysis unit performs the morphological analysis after performing a process of replacing a spoken word included in the sentence with a sentence word.
Thereby, the precision of morphological analysis can be improved.

Furthermore, it is preferable that the user preference analysis unit performs a process of deleting at least personal information, discriminatory terms, and secret words from each morpheme after performing the morpheme analysis.
Thereby, the safety of the service can be ensured.

The information processing apparatus according to the present invention includes a preference similarity determination unit that determines whether or not the user and other users have similar preferences based on information stored in the user preference information storage unit, and the preference The similarity determination unit includes a user information providing unit that provides the user with information specifying other users determined to have close preferences.
Thus, the user can meet other users who have similar preferences with the user on the network, and can avoid unintentional contact with other users who do not meet the preferences.

In addition, the preference similarity determination unit includes at least what nouns the user frequently uses, what nouns are used in a positive sense, and what nouns are negative values. The user's preference information vector is created based on the information on whether the user is using the meaning, and the similarity between the user's preference information vector and the other user's preference information vector is calculated. It is desirable to determine whether or not is close.
Thereby, the closeness of preference can be obtained as a numerical value by a simple method.

The user information providing unit sets the other user's email address determined to be close by the preference similarity determination unit to the destination of the email that is assumed to be sent to an unspecified user. User information can be provided to the user.
Thereby, since the mail assumed to be sent to an unspecified user reaches other users with similar preferences, it is possible to create an opportunity to get to know other users with similar preferences.

The user information providing unit displays a pet character of another user determined to be close in the preference similarity determination unit on the terminal device of the user, and the information on the other user is displayed via the pet character. Can be provided.
As a result, it is possible to create an opportunity to get acquainted with other users who have similar tastes efficiently.

In addition, the preference similarity determination unit determines that the user's answer to the question for determining whether or not the user's preference is close is as expected only when the user's answer is as expected. You may make it determine with preference with the said other user.
Thereby, it is possible to provide information on other users who are closer in taste.

The information processing apparatus according to the present invention includes an automatic sentence creation unit that automatically creates a sentence using nouns registered in the user preference information storage unit, and a sentence created by the sentence automatic creation unit. Is provided to the user via the user's pet character.
As a result, the user is concerned about the content of the text provided by the pet and frequently logs into the site.

The sentence providing unit may provide the sentence created by the sentence automatic creating unit via the pet in a conversation between the user and the pet via the input unit and the output unit of the terminal device. it can.
The sentence providing unit may provide the sentence created by the sentence automatic creating unit by the pet of the user writing on a diary or bulletin board.
Furthermore, the sentence providing unit may provide the sentence created by the sentence automatic creating unit by an electronic mail transmitted by the user's pet to the user.

  The program according to the present invention is a program for connecting a computer to a plurality of user terminal devices via a communication network and executing a function of providing a communication service between the users, wherein the user performs the terminal device. The function of receiving the text data input via the communication network, and analyzing the text, the noun in the text is used in a positive or negative meaning And the function of registering in the user preference information storage unit is executed.

  The program of the present invention is stored in a storage medium and distributed. Such a storage medium is a computer-readable medium such as various ROMs, USB memories equipped with flash memories, SD memories, memory sticks, memory cards, FDs, CD-ROMs, DVD-ROMs, etc. A typical storage medium. In addition, the broad recording medium includes a transmission medium such as the Internet capable of transmitting the program. This is because the program (downloaded) transmitted through the transmission medium is stored in the memory as it is and executed by the computer.

  ADVANTAGE OF THE INVENTION According to this invention, the service which is safe for a user based on a user's preference and cannot get tired can be provided.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
Embodiment 1 FIG.
FIG. 1 is a block diagram showing an overall configuration of an information processing system 100 including a service providing server (information processing apparatus) 10 according to the first embodiment of the present invention. As illustrated in FIG. 1, the information processing system 100 includes a service providing server 10 and a user terminal device 20 connected to a network 500.

  The service providing server 10 is an information processing apparatus that provides a social networking service (SNS) to a registered member. The service providing server 10 may be composed of a single computer or may be composed of a plurality of computers distributed on a network.

  The user terminal device 20 is, for example, a personal computer (PC), a CPU as a control means for controlling the operation and processing of each unit, a memory functioning as a work area for storing various information and processing, a storage such as a hard disk device, and the like. Means, an input interface for receiving signals input from input means such as a keyboard and a mouse, an output interface for outputting signals to output means such as a display device, a communication interface for transmitting and receiving information connected to the network 500, and Provide a bus to connect. The user terminal device 20 may be a personal computer (PC), a personal digital assistant (PDA), a mobile phone, or the like.

  The network 500 is a communication line for transmitting and receiving information between the service providing server 10 and the user terminal device. For example, it may be any of the Internet, a LAN, a dedicated line, a packet communication network, a telephone line, a corporate network, other communication lines, combinations thereof, and the like, regardless of whether they are wired or wireless.

  FIG. 2 is a block diagram illustrating a configuration of the service providing server 10 according to the present embodiment. As shown in FIG. 2, the service providing server 10 includes a control unit 11 for controlling the operation and processing of each unit, a storage unit 12 for storing information, and a communication interface for transmitting and receiving information via a network. Communication unit 13 and a bus 14 connecting them. The control unit 11 is a processor such as a CPU, for example. The storage unit 12 is, for example, a memory such as a ROM or a RAM, or a storage device such as a hard disk device.

  The control unit 11 includes an input information reception unit 111, a user preference analysis unit 112, a preference similarity determination unit 113, another user information provision unit 114, an automatic sentence creation unit 115, a pet control unit 116, a member registration unit 117, and a service provision unit. 118.

  The storage unit 12 includes a member information storage unit 121, a user preference information storage unit 122, a noun category information storage unit 123, a hieroglyph conversion information storage unit 124, a spoken language conversion information storage unit 125, and an NG word information storage unit 126. ing.

  The input information receiving unit 111 is a network that stores information input by the user using the input unit of the user terminal device 20, such as writing to a diary or bulletin board, contents of an e-mail, and contents returned in response to a question from a pet. Receive via 500.

  The user preference analysis unit 112 analyzes the input to the diary or bulletin board received by the input information receiving unit 111, the content of the e-mail, the content returned in response to the question from the pet, and the noun in the sentence is positive Whether it is used with a negative meaning or a negative meaning, and registers in the user preference information storage unit 122.

  The preference similarity determination unit 113 determines whether or not the preference with other users is close based on information registered in the user preference information storage unit 122.

  The other user information providing unit 114 provides the user with information (member ID, e-mail address, etc.) that specifies the other user whose preference is determined to be close by the preference similarity determination unit 113.

  The automatic sentence creation unit 115 automatically creates sentences using nouns registered in the user preference information storage unit 122.

  The pet control unit 116 outputs the sentence created by the sentence automatic creating unit 115 via the pet. Here, a pet is a character on the SNS that exchanges messages with the user or writes a diary or bulletin board on behalf of the user based on the member information and preference information of the user.

  The member registration unit 117 has a function of registering a new user as a member. In the new member registration process, a new member ID is issued, and the member ID is registered in the member information storage unit 121 in association with the member information including the new user's e-mail address, password, and profile information.

  The service providing unit 118 has a function to provide SNS members, that is, members registered in the member information storage unit 121, authentication processing at the time of SNS login and various services (diary, community, etc.) related to the SNS. Have.

  The member information storage unit 121 stores member information of members registered in the SNS. The member information includes a member ID, a password, an e-mail address, a mobile mail address, a surname / first name, a nickname, an address, gender, self-introduction, a mobile phone terminal ID, and the like.

  The user preference information storage unit 122 stores information analyzed by the user preference analysis unit 112 as to whether a noun in a sentence is used in a positive or negative meaning.

  The noun category information storage unit 123 stores information on the category to which the noun belongs. The category includes a small category and a large category. All registered nouns belong to one or more small categories, and each small category belongs to one or more large categories. For example, “diamond” and “carbon” belong to the same small category, and “archery” and “archery” belong to the same small category. One noun may belong to a plurality of categories, for example, “car racing” belongs to both a category in which car-related nouns are collected and a category in which sports-related nouns are collected.

  The hieroglyph conversion information storage unit 124 stores information for replacing characters, pictograms, approximate characters, and the like included in the sentence with normal characters.

  The spoken word conversion information storage unit 125 stores information for replacing a spoken word contained in a sentence with a sentence word.

  In the NG word information storage unit 126, personal information to be excluded from sentences, discrimination terms, secret words, and the like are registered.

  Next, the user preference information analysis process in the service providing server 10 will be described using the flowchart of FIG.

  First, the user A logs in to the SNS provided by the service providing server 10 via the network 500 from the user terminal device 20. At this time, the user A inputs the member ID and password using the input means of the user terminal device 20, and the service providing unit 118 of the service providing server 10 stores the member ID and password received via the network 500 as member information. If it is registered in the unit 121, a login process is performed, and the main screen is displayed on the output means of the user terminal device 20 (step S11).

  Next, when the user A uses the input means of the user terminal device 20 to write a diary or bulletin board, send an e-mail, or have a conversation with a pet, the input information receiving unit 111 of the service providing server 10 writes The information of the received contents is received via the network 500 (step S12).

  Next, the user preference analysis unit 112 of the service providing server 10 analyzes the written content received by the input information reception unit 111, extracts nouns used in a positive sense in the sentence, and stores user preference information. Register in the unit 122 (step S13).

  The sentence analysis process by the user preference analysis unit 112 will be described using the flowchart of FIG.

  First, the user preference analysis unit 112 performs character normalization processing in step S21. Specifically, when the input sentence includes alphanumeric characters, all of them are converted into half-width characters. In addition, when half-tone katakana characters with muddy sounds and semi-voiced sounds are included, the character changes to one full-width character.

  Next, hieroglyph conversion processing is performed in step S22. This is a process of replacing characters, pictograms, approximate characters, etc. used by young people with ordinary characters. As an example, “OLILI” is converted to “delicious”, “Nani no Ishii” is converted to “fun”, and “Ganbatsu τ” is converted to “Good luck”. The user preference analysis unit 112 performs conversion processing with reference to the hieroglyph conversion information storage unit 124. Thereby, analysis accuracy can be raised more.

  Next, colloquial conversion processing is performed in step S23. By converting spoken language into written words, the accuracy of morphological analysis to be performed later can be improved. For example, “good” is converted to “good”, and “Magico” is converted to “really good”. The user preference analysis unit 112 performs conversion processing with reference to the colloquial conversion information storage unit 125.

  Next, in step S24, the user preference analysis unit 112 performs a morphological analysis of the sentence input by the user A, decomposes the sentence into morpheme units (the smallest unit having meaning), and determines the part of speech of each unit. The morphological analysis can be performed using a known morphological analysis tool.

  In step S25, NG word elimination processing is performed. For example, items related to personal information such as telephone numbers and email addresses, discriminatory terms, secret words, etc. are deleted from the text. The user preference analysis unit 112 refers to the NG word information storage unit 126 and deletes the corresponding NG word.

  Next, in step S26, the user preference analysis unit 112 analyzes the dependency of each word decomposed into morpheme units in step S25, so that a noun is used in a positive meaning or a negative meaning. It is determined whether or not it is used. In this case, the relationship between the words in the subject predicate relationship does not affect the determination, and is determined by analyzing the relationship of modification or connection.

  For example, in the sentence “pasta: ha: wheat flour :: made”, “pasta” and “made” do not affect the judgment because they are related to the subject predicates. On the other hand, in the sentence “Delicious: Pasta: Ha: Flour: Made: Made”, “Pasta” and “Delicious” are subject to analysis because they have a modification relationship. In this case, since the modification relationship has a positive meaning, it is determined that “pasta” is used in a positive sense in this sentence.

  Also, for example, in the sentence “Especially: no: Pasta: is: ruined”, “pasta” and “messed up” are the relationship between the subject predicates, and therefore do not affect the judgment. On the other hand, in the sentence “Pasta: is: entered: was: because: ruined”, “pasta” and “messed up” are connected, so in this case, the connection relationship has a negative meaning. Therefore, in this sentence, it is determined that “pasta” is used in a negative sense.

  Also, in the sentence "Pasta: is: entered: did not: so: messed up", "pasta" and "messed up" are connected, and in this case they are used in double negation, In this sentence, it is determined that “pasta” is used in a positive sense.

  In the case of the text “beautiful: watermill: no: maiden”, it cannot be analyzed whether “beautiful” relates to “watermill” or “maiden”, so it is not subject to judgment.

  As described above, the determination result executed in step S26 is stored in the user preference information storage unit 122 (step S27). FIG. 5 is a diagram illustrating an example of information stored in the user preference information storage unit 122. In the figure, “category ID” is an ID of a small category to which each noun belongs, and this refers to, for example, a small category dictionary stored in the noun category information storage unit 123 as shown in FIG. Is obtained by “Interest” in the figure is incremented by 1 point each time a noun is analyzed in the sentence analysis of the user preference analysis unit 112. In addition, “positive value” is added when a noun that is a result of sentence analysis is positively used. As for the “negative value”, points are added when a noun is negatively used as a result of sentence analysis. It should be noted that “positive value” and “negative value” may be added at different points depending on the type of sentence (diary, bulletin board, e-mail, answer to a pet question, etc.) that has been analyzed.

  Next, a process for introducing another user who has a similar preference to the user A as a friend based on the information registered in the user preference information storage unit 122 as described above will be described.

  The preference similarity determination unit 113 determines whether the preference with other users is close based on information registered in the user preference information storage unit 122. Specifically, after creating a preference information vector by extracting one or more categories from the user preference information storage unit 122 in accordance with the purpose of a category having a high “interest”, “positive value”, or “negative value” point This is done by calculating the similarity with other users' preference information vectors.

  As a method of creating a preference information vector, for example, an item corresponding to the dimension of the vector is determined in advance, and a calculation rule for determining the value of each dimension according to the extracted points for each category is provided. Can be used. At this time, the small category registered in the user preference information storage unit 122 may be replaced with a large category that is further classified. For example, as shown in FIG. 7, a correspondence table of small categories belonging to each large category is stored in the noun category information storage unit 123, and the extracted small categories are replaced with large categories. Calculate points. Items corresponding to the dimension of the vector are determined in advance as shown in FIG. 8, for example, and the value of each item is calculated based on the points for each large category. An item value calculation method is determined based on a predetermined calculation rule.

  With the above method, preference information vectors of arbitrary other users are created, and an angle formed with the preference information vector of user A is calculated. As a result of the calculation, the user is defined as a user having a close preference in order from the other user whose angle between the vectors is small.

  And the other user information provision part 114 provides the user's member ID and the information of an e-mail address to the user A in an order from the other user determined that the preference similarity determination part 113 has the closest preference. In addition, information such as first name, last name, nickname, gender, and self-introduction may be provided at the same time.

  The method of providing user information with similar preferences may directly present information such as the member ID and e-mail address of the corresponding user. For example, e-mail that is supposed to be sent to an unspecified user (in the real world, message It is also possible to automatically set the address of a user who is determined to be close to the destination of the transmission destination. Alternatively, the pet B of the user B determined to be close in preference may be displayed on the user terminal device 20 of the user A, and information on the user B having close preference may be provided via the pet b.

  In addition to the calculation using the above vector, a condition for the preference similarity determination unit 113 to determine whether or not other users' preferences are close may be further provided. For example, a question for determining whether or not the preference is close may be set in advance and introduced to the user A only when the other user's answer to the question is as expected. . Note that this question exchange may be performed by pets between users.

  Next, based on the information registered in the user preference information storage unit 122, the user A's pet changes to the user A and writes in a diary or bulletin board, or sends an email to the user A Will be described.

  The automatic sentence creation unit 115 automatically creates a sentence using nouns registered in the user preference information storage unit 122. For example, when the user preference information storage unit 122 is referred to and the user A frequently uses the noun “pasta” in a positive sense, the automatic sentence creation unit 115 displays the topic sentence regarding “pasta”. create.

  Further, the pet control unit 116 outputs the text created by the text automatic creation unit 115 via the user A's pet. The prepared text is provided, for example, by a conversation with a pet on SNS, a pet writes in a diary or bulletin board of user A, or a pet sends an email to user A. The By doing so, the user A is worried about the writing of the pet and the content of the mail, and frequently logs in to the SNS.

  As described above, according to the present embodiment, based on the analysis results such as writing to the user's diary or bulletin board, the contents of the e-mail, the contents returned to the question from the pet, etc. A user preference information storage unit 122 that stores information on what nouns are used positively and what nouns are used negatively, and introduces other users with similar preferences, or pets are users Since the user's interest is written or an e-mail is sent, it is possible to provide a service that is safe and secure for the user based on the user's preference.

  Further, since the contents of the user preference information storage unit 122 are constantly updated every time the user writes, it is possible to use what the user has recently been interested in and preferences in a timely manner.

  The present invention is not limited to the above-described embodiment, and can be implemented in various other forms without departing from the gist of the present invention. For this reason, the said embodiment is only a mere illustration in all points, and is not interpreted limitedly. For example, the above-described processing steps can be executed in any order or in parallel as long as there is no contradiction in the processing contents.

  Moreover, although this embodiment demonstrated as what the service provision server 10 provides SNS, it is applicable not only when it is SNS but when providing the communication service of the users via a communication network.

It is a block diagram which shows the whole structure of the information processing system containing the service provision server (information processing apparatus) by Embodiment 1 of this invention. It is a block diagram which shows the structure of the service provision server by Embodiment 1 of this invention. It is a flowchart of the user preference information analysis process in the service provision server by Embodiment 1 of this invention. It is a flowchart of the text analysis process by the user preference analysis part by Embodiment 1 of this invention. It is a figure which shows the example of the information memorize | stored in a user preference information storage part. It is a figure which shows the example of the category dictionary memorize | stored in the noun category information storage part. It is a figure which shows the example of the correspondence table of the large category memorize | stored in the noun category information storage part and a small category. It is a figure which shows the example of a preference information vector.

Explanation of symbols

  10 service providing server, 20 user terminal device, 100 information processing system, 500 network, 11 control unit, 12 storage unit, 13 communication unit, 14 bus, 111 input information receiving unit, 112 user preference analysis unit, 113 preference similarity determination 114, other user information providing unit, 115 automatic sentence creation unit, 116 pet control unit, 117 member registration unit, 118 service providing unit, 121 member information storage unit, 122 user preference information storage unit, 123 noun category information storage unit, 124 hieroglyph conversion information storage unit, 125 colloquial conversion information storage unit, 126 NG word information storage unit

Claims (16)

An information processing apparatus that is connected to a plurality of user terminal devices via a communication network and provides a communication service between each user,
An input information receiving unit that receives text data input by the user via the terminal device via the communication network;
A user preference analysis unit that analyzes the sentence, analyzes whether a noun in the sentence is used in a positive or negative sense, and registers the noun in a user preference information storage unit; Information processing equipment. The user preference analysis unit
By analyzing the morphological analysis of the sentence, the nouns contained in the sentence are discriminated, and whether each noun is used in the sentence in a positive or negative meaning is analyzed. The information processing apparatus according to claim 1. The user preference analysis unit
By analyzing the meaning of a morpheme that is modified or connected to a noun in the sentence, the noun is used in a positive or negative meaning in the sentence. The information processing apparatus according to claim 2, wherein the information processing apparatus is analyzed. The user preference analysis unit
The information processing according to claim 2, wherein the morphological analysis is performed after performing a process of replacing at least a hit character, a pictogram, and an approximate character included in the sentence with normal characters. apparatus. The user preference analysis unit
The information processing apparatus according to any one of claims 2 to 4, wherein the morphological analysis is performed after performing a process of replacing a spoken word contained in the sentence with a sentence word. The user preference analysis unit
The information according to any one of claims 2 to 5, wherein after the morpheme analysis is performed, processing for deleting at least personal information, discriminatory terms, and secret words from each morpheme is performed. Processing equipment. A preference similarity determination unit that determines whether or not the user and other users have similar preferences based on information stored in the user preference information storage unit;
The user preference providing unit according to any one of claims 1 to 6, further comprising: a user information providing unit that provides the user with information specifying another user who is determined to have close preferences in the preference similarity determination unit. The information processing apparatus described in 1. The preference similarity determination unit
At least one of the nouns frequently used by the user, what nouns are used in a positive sense, and what nouns are used in a negative sense Based on the information, creating a preference information vector of the user and calculating a similarity with the preference information vector of another user to determine whether the preference between the user and the other user is close The information processing apparatus according to claim 7. The user information providing unit
By setting the mail address of another user determined to be close in the preference similarity determination unit as the mail destination assumed to be sent to an unspecified user, the information on the other user is sent to the user. The information processing apparatus according to claim 7, wherein the information processing apparatus is provided. The user information providing unit
The pet character of another user determined to be close in preference by the preference similarity determination unit is displayed on the terminal device of the user, and the information of the other user is provided to the user via the pet character. The information processing apparatus according to any one of claims 7 to 9. The preference similarity determination unit
It is determined that the preference between the user and the other user is close only when the other user's answer to the question for determining whether or not the user's preference is close is as expected. The information processing apparatus according to claim 7, wherein the information processing apparatus is an information processing apparatus. A sentence automatic creation unit that automatically creates a sentence using nouns registered in the user preference information storage unit;
The information according to claim 1, further comprising: a sentence providing unit that provides the user with the sentence created by the automatic sentence creating unit via the pet character of the user. Processing equipment. The sentence providing unit
The sentence created by the sentence automatic creation unit is provided via the pet in a conversation between the user and the pet via the input unit and the output unit of the terminal device. Information processing device. The sentence providing unit
The information processing apparatus according to claim 12, wherein the user's pet provides a sentence created by the sentence automatic creation unit by writing on a diary or a bulletin board. The sentence providing unit
The information processing apparatus according to claim 12, wherein the user's pet provides the sentence created by the sentence automatic creation unit by an electronic mail transmitted to the user. On the computer,
A program for connecting a plurality of user terminal devices via a communication network and executing a function of providing a communication service for each user,
A function for receiving text data input by the user via the terminal device via the communication network;
The program which analyzes the said sentence, analyzes whether the noun in a sentence is used by the positive meaning or the negative meaning, and performs the function registered into a user preference information storage part.

Images