Information processing device, information processing method, and program

Abstract

An information processing device, an information processing method, and a program for improving the effectiveness of advertisements on a community site are provided. [Solution] The information processing device 1 has an acquisition unit 131 that acquires behavioral information indicating the behavior of a first user, an identification unit 132 that identifies a second user who exhibits behavior similar to that of the first user, an evaluation unit 133 that inputs the behavioral information of the second user into a behavior learning model and outputs an estimated value indicating the degree to which the user behavior indicated by the behavioral information is estimated to be the behavior of a user belonging to the community, an extraction unit 134 that extracts behavioral information of the second user to be displayed on the information terminal of the first user based on the estimated value output by the evaluation unit, and a display control unit 135 that displays the behavioral information extracted by the extraction unit on the information terminal of the first user and controls information for viewing information regarding a specified product to be associated with the behavioral information extracted by the extraction unit and further displayed on the information terminal of the first user.

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Advertisements are displayed in social networking services (SNS). For example, an advertisement system is known that encourages users to post on SNS containing advertisements (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2018-195026 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, in conventional technology, the reward was only awarded if a user posted content specified by the advertiser, which resulted in the problem that such posts had little appeal to viewers.

[0005] The present invention has been made in consideration of these points, and has an object to improve the effectiveness of advertising on community sites. [Means for solving the problem]

[0006] In a first aspect of the information processing device of the present invention, the information processing device has a memory unit that stores a behavioral learning model that has learned behavioral information of each of a plurality of users participating in a specified community, the behavioral learning model being trained to receive behavioral information indicating a user's behavior and output an estimated value indicating the degree to which the user's behavior indicated by the input behavioral information is estimated to be the behavior of a user belonging to the community; an acquisition unit that acquires behavioral information indicating a first user participating in the community; an evaluation unit that, upon the acquisition unit acquiring the behavioral information of the first user, inputs behavioral information of a second user different from the first user into the behavioral learning model and outputs an estimated value; an extraction unit that extracts behavioral information of the second user based on the estimated value output by the evaluation unit to determine information to be displayed on the information terminal of the first user; a determination unit that determines information to be displayed on the information terminal of the first user based on the content of the behavioral information extracted by the extraction unit; and a display control unit that controls the information determined by the determination unit to be displayed on the information terminal of the first user.

[0007] The memory unit may further store a general-purpose learning model which is a pre-trained general-purpose language model and is configured to input behavioral information and output an estimated value, the behavioral learning model being a trained model obtained by fine-tuning the general-purpose learning model based on behavioral information of a plurality of users participating in the specified community, the evaluation unit may input the behavioral information into the general-purpose learning model and further output an estimated value, and the extraction unit may extract behavioral information to be displayed on the information terminal of the first user based on the estimated value that the evaluation unit has caused the behavioral learning model to output and the estimated value that the evaluation unit has caused the general-purpose learning model to output.

[0008] The extraction unit may extract behavioral information to be displayed on the information terminal of the first user, the behavioral information being such that the magnitude relationship between the estimated value that the evaluation unit has output to the behavioral learning model and the estimated value that the evaluation unit has output to the general-purpose learning model satisfies a predetermined condition.

[0009] The extraction unit may convert each of the first user's behavioral information and the second user's behavioral information into vectors, calculate a distance between the vector corresponding to the second user's behavioral information and the vector corresponding to the first user's behavioral information, and extract behavioral information for which the calculated distance is equal to or greater than a predetermined threshold as behavioral information to be displayed on the information terminal of the first user.

[0010] The display control unit may control the behavioral information extracted by the extraction unit to be displayed on the information terminal of the first user, and may also control the information determined by the determination unit to be associated with the behavioral information extracted by the extraction unit and displayed on the information terminal of the first user.

[0011] The behavioral information is text data indicating user comments in a community, and the display control unit may further display the behavioral information extracted by the extraction unit on the information terminal of the first user, and if the behavioral information extracted by the extraction unit includes a string indicating a specified product, may control information for viewing information regarding the specified product to be associated with the behavioral information extracted by the extraction unit and displayed on the information terminal of the first user.

[0012] The display control unit may further display the behavioral information extracted by the extraction unit on the information terminal of the first user, and when the behavioral information extracted by the extraction unit indicates a behavior related to a specified product, display information for viewing information about the specified product on the information terminal of the first user in association with the behavioral information extracted by the extraction unit, and the information processing device may further have a benefit granting unit that grants a specified benefit to the second user when the first user purchases the specified product via the information for viewing information about the specified product displayed on the information terminal by the display control unit.

[0013] In the information processing method according to the second aspect of the present invention, an acquisition step of acquiring behavior information indicating the behavior of a first user who participates in a predetermined community, which is executed by a computer, and in the acquisition step, upon the acquisition of the behavior information of the first user, a behavior learning model that has learned the behavior information of each of a plurality of users who participate in the predetermined community, which is stored in a storage unit, and that is learned to take behavior information as an input and output an estimated value indicating the degree to which the behavior of the user indicated by the input behavior information is estimated to be the behavior of a user belonging to the community, an evaluation step of inputting the behavior information of a second user different from the first user to the behavior learning model and outputting an estimated value, an extraction step of extracting the behavior information of the second user for determining the information to be displayed on the information terminal of the first user based on the estimated value output in the evaluation step, a determination step of determining the information to be displayed on the information terminal of the first user based on the content of the behavior information extracted in the extraction step, and a display control step of further displaying the information determined in the determination step on the information terminal of the first user.

[0014] In the program according to the third aspect of the present invention, the computer is caused to execute an acquisition step of acquiring behavior information indicating the behavior of a first user who participates in a predetermined community, an evaluation step of inputting the behavior information of a second user different from the first user to a behavior learning model that has learned the behavior information of each of a plurality of users who participate in the predetermined community, which is stored in a storage unit, and that is learned to take behavior information as an input and output an estimated value indicating the degree to which the behavior of the user indicated by the input behavior information is estimated to be the behavior of a user belonging to the community, and upon the acquisition of the behavior information of the first user in the acquisition step, outputting an estimated value, an extraction step of extracting the behavior information of the second user for determining the information to be displayed on the information terminal of the first user based on the estimated value output in the evaluation step, and a determination step of determining the information to be displayed on the information terminal of the first user based on the content of the behavior information extracted in the extraction step. A display control step of further displaying the information determined in the determination step on the information terminal of the first user is executed.

[0015]

[0016] Effect of the Invention

[0017] According to the present invention, it is possible to improve the effectiveness of advertisements on a community site. [Brief description of the drawings]

[0018] [Figure 1] 1 is a diagram for explaining an overview of an information processing system S according to an embodiment. [Diagram 2] 1 is a block diagram showing a configuration of an information processing device 1. FIG. [Diagram 3] 4 is a diagram showing an example of a behavior information table stored in the storage unit 12. FIG. [Figure 4] FIG. 11 is a diagram illustrating an example of a process performed by an evaluation unit 133. [Diagram 5] 13 is a diagram showing an example of a screen displayed by a display control unit 135. FIG. [Figure 6] 3 is a flowchart showing a process flow in the information processing device 1. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] [Outline of Information Processing System S] FIG. 1 is a diagram for explaining an overview of an information processing system S according to an embodiment. The information processing system S is a system for managing a fan community (hereinafter, sometimes referred to as a "predetermined community"). A fan community is a community for users who are fans of a particular product, service, brand, hobby, person, organization, group, etc. to interact with each other. More specifically, a fan community is an SNS in which participating users post images, text, etc., view other users' posts, rate them, comment on them, etc., and thereby users interact with each other. As an example, an information processing device 1 includes the information processing system S, which includes an information processing device 1 and an information terminal 2.

[0020] The information processing device 1 is a device that manages a fan community. The information processing device 1 may manage each of a plurality of communities set for each theme. The information processing device 1 accepts posts from users' information terminals 2, and causes the information terminals 2 to display information posted to the communities.

[0021] The information terminal 2 is a terminal used by a user. The information terminal 2 is a smartphone, a tablet, or a personal computer. The information terminal 2 transmits, for example, posts made by the user to a community to the information processing device 1, and displays information acquired from the information processing device 1.

[0022] The processing in the information processing system S will be described. The information processing device 1 acquires behavioral information of a first user. The behavioral information is information indicating the behavior of the user. Examples of the behavioral information include text posted by the user on an SNS, a product purchase history, or a log of a device used by the user. The first user is a user who is to determine the information to be displayed.

[0023] The information processing device 1 identifies a user who exhibits behavior similar to the behavior of the first user as a second user based on the acquired behavior information of the first user ((1) in FIG. 1). As an example, the information processing device 1 calculates a similarity between the behavior information of the first user and the behavior information of other users who belong to the community, and identifies the second user based on the calculated similarity.

[0024] The information processing device 1 extracts behavioral information to be displayed on the information terminal 2 of the first user from the behavioral information of the second user ((2) in FIG. 1). The information processing device 1 inputs the behavioral information of the second user to the behavior learning model, causes it to output an estimated value, and extracts the behavioral information based on the estimated value output by the behavior learning model. The behavior learning model is a trained model that has learned the behavioral information of each of a plurality of users participating in a specified community. The behavior learning model is trained to receive behavioral information as input and output an estimated value corresponding to the input behavioral information. The estimated value is a value indicating the degree to which the behavior of the user indicated by the behavior information is estimated to be the behavior of a user belonging to the community. As an example, the information processing device 1 extracts behavioral information whose estimated value is equal to or greater than a specified threshold value.

[0025] The information processing device 1 displays the extracted behavioral information on the information terminal 2 of the first user ((3) in FIG. 1). When the extracted behavioral information indicates behavior related to a predetermined product, the information processing device 1 further displays information for viewing information related to the predetermined product on the information terminal 2 of the first user. The predetermined product is, for example, a product or service related to a service, brand, hobby, person, organization, group, etc. that is the target of a predetermined community. The information processing device 1 may display information for viewing information related to a service, brand, store, character, person, organization, group, etc. related to the community, instead of information related to the product. As an example, the information processing device 1 displays a screen including a link for accessing information related to the predetermined product on the information terminal 2 of the first user.

[0026] By configuring the information processing device 1 in this way, it is possible to improve the effectiveness of advertisements on the community site.

[0027] [Configuration of information processing device 1] 2 is a block diagram showing the configuration of the information processing device 1. The information processing device 1 has a communication unit 11, a storage unit 12, and a control unit 13. The control unit 13 has an acquisition unit 131, an identification unit 132, an evaluation unit 133, an extraction unit 134, a display control unit 135, a calculation unit 136, a determination unit 137, and a benefit granting unit 138.

[0028] The communication unit 11 is a communication interface for transmitting and receiving data to and from other devices via a network. The storage unit 12 is a storage medium including a ROM (Read Only Memory), a RAM (Random Access Memory), an SSD (Solid State Drive), a hard disk drive, etc. The storage unit 12 stores in advance a program to be executed by the control unit 13.

[0029] The storage unit 12 stores behavioral information indicating user behavior, which is behavioral information for each of a plurality of users participating in a predetermined community. FIG. 3 is a diagram showing an example of a data structure of a behavioral information table stored in the storage unit 12. In the behavioral information table shown in FIG. 3, an "behavior ID," a "destination community," a "posting user," and "behavior information" are associated with each other. The "behavior ID" is an ID (Identification) for identifying behavioral information. The "destination community" is an ID indicating the community to which the behavioral information was posted. The "posting user" indicates the user ID of the user who posted the behavioral information. The "behavioral information" is information such as text, images, and videos posted by a user to a community. In the behavioral information table, an estimated value corresponding to the behavioral information may be further associated.

[0030] The storage unit 12 stores the behavioral learning model. The storage unit 12 may store user information in which a user ID for identifying a user is associated with information indicating a community to which the user belongs.

[0031] The control unit 13 is a processor such as a CPU (Central Processing Unit), etc. The control unit 13 executes the programs stored in the storage unit 12 to function as an acquisition unit 131, an identification unit 132, an evaluation unit 133, an extraction unit 134, a display control unit 135, a calculation unit 136, a determination unit 137, and a benefit granting unit 138.

[0032] The acquisition unit 131 acquires behavioral information indicating the behavior of a first user participating in a community. The behavioral information includes, for example, a user ID of a user who posts, a community ID indicating a community to which the post is to be posted, and the content to be posted. For example, the behavioral information is text data indicating a statement made by the user in the community. The acquisition unit 131 acquires behavioral information of a user to be determined from the information terminal 2.

[0033] The identification unit 132 calculates the similarity between the behavioral information of the first user and the behavioral information of each of the multiple users stored in the storage unit 12, and identifies a second user who exhibits behavior similar to the behavior of the first user based on each calculated similarity. As an example, the identification unit 132 converts the behavioral information of the first user and the behavioral information of the multiple users into vectors. The identification unit 132 calculates the similarity between a vector based on the behavioral information of the first user and a vector based on the behavioral information of the second user. An example of the similarity is cosine similarity, but is not limited to this. As an example, the identification unit 132 identifies a user corresponding to the behavioral information with the largest calculated similarity as the second user.

[0034] A user whose behavioral information has a similarity equal to or greater than a predetermined threshold may be identified as a second user. That is, the identification unit 132 calculates the similarity between the behavioral information of the first user and the behavioral information of multiple users stored in the storage unit 12, and identifies a user whose behavioral information has a similarity equal to or greater than a predetermined threshold with respect to the behavioral information of the first user as a second user.

[0035] The evaluation unit 133 inputs the behavioral information of the second user into the behavioral learning model and outputs an estimated value. The evaluation unit 133 refers to the storage unit 12, acquires the behavioral information of the second user identified by the identification unit 132, inputs the acquired behavioral information into the behavioral learning model, and outputs an estimated value. As an example, the evaluation unit 133 refers to the storage unit 12, inputs one or more pieces of behavioral information associated with the second user into the behavioral learning model, and outputs an estimated value for each of the one or more pieces of behavioral information.

[0036] An example of the processing of the evaluation unit 133 will be described with reference to FIG. 4. As an example, the evaluation unit 133 breaks down the behavioral information A1 of the second user into tokens (T1 to T6). A token is the smallest unit processed as natural language, and is a character, a word, or a vector corresponding to the contents indicated by the purchase history of a specific product or a device log. The evaluation unit 133 inputs the tokens generated by breaking down the behavioral information into a behavioral learning model in an autoregressive manner, and outputs an estimated value. In this case, the estimated value is a conditional probability that the input token will be consecutive when a previously input token is given. As an example, the evaluation unit 133 outputs the sum of the conditional probabilities output by the behavioral learning model for each token as an estimated value.

[0037] The extraction unit 134 extracts behavioral information of the second user to be displayed on the information terminal of the first user based on the estimated value output by the evaluation unit 133. As an example, the extraction unit 134 extracts behavioral information having an estimated value equal to or greater than a threshold value from among the behavioral information associated with the second user identified by the identification unit 132. The extraction unit 134 may extract behavioral information having the largest estimated value from among the behavioral information associated with the second user identified by the identification unit 132.

[0038] The display control unit 135 displays the behavioral information extracted by the extraction unit 134 on the information terminal of the first user. When the behavioral information extracted by the extraction unit 134 indicates a behavior related to a predetermined product, the display control unit 135 controls to further display information for viewing information about the predetermined product on the information terminal of the first user in association with the behavioral information extracted by the extraction unit 134. More specifically, when the behavioral information is text that mentions the predetermined product, the display control unit 135 displays information about the predetermined product. That is, the display control unit 135 displays the behavioral information extracted by the extraction unit 134 on the information terminal of the first user, and when the behavioral information extracted by the extraction unit 134 includes a character string indicating the predetermined product, the display control unit 135 controls to display information for viewing information about the predetermined product on the information terminal of the first user in association with the behavioral information extracted by the extraction unit 134.

[0039] Fig. 5 is a diagram showing an example of a screen displayed by the display control unit 135. In the screen shown in Fig. 5, an object OB for viewing information related to a product associated with the behavioral information is displayed together with the behavioral information extracted by the extraction unit 134. A link for viewing information related to the product is embedded in the object OB. When the user operates the information terminal 2 and presses the object OB displayed on the screen, a page showing information related to the product is displayed on the information terminal 2.

[0040] As an example, the storage unit 12 stores product information in which a product name is associated with a link for viewing information related to the product. The product information may further be associated with an ID for identifying the product. When the behavioral information includes a character string indicating a product, the display control unit 135 causes the information terminal 2 to display a screen including a link in the product information that corresponds to the character string indicating the product included in the behavioral information.

[0041] By configuring the information processing device 1 in this way, it is possible to improve the effectiveness of advertisements on the community site.

[0042] The information processing device 1 may be configured to determine behavioral information to be extracted by comparing with an estimated value output by a general-purpose learning model. In this case, the behavioral learning model is a trained model in which the general-purpose learning model is fine-tuned based on behavioral information of a plurality of users participating in a specific community. The general-purpose learning model is a pre-trained general-purpose language model. The general-purpose learning model is trained to be able to execute natural language processing based on a large amount of data set. The general-purpose learning model is configured to output an estimated value using behavioral information as input. The storage unit 12 stores the general-purpose learning model.

[0043] Compared to general-purpose learning models, behavioral learning models are trained using behavioral information of users belonging to a specific community as training data, and therefore differ from general-purpose learning models in that when behavioral information of a user belonging to a specific community is input, it is expected to output degree information indicating a higher likelihood that the user belongs to the specific community.

[0044] The evaluation unit 133 inputs the behavioral information to the general-purpose learning model and outputs an estimated value. The evaluation unit 133 inputs the behavioral information to be input to the behavioral learning model to the general-purpose learning model. The evaluation unit 133 inputs the behavioral information of the second user identified by the identification unit 132 to the general-purpose learning model and outputs an estimated value. The evaluation unit 133 refers to the storage unit 12, inputs one or more pieces of behavioral information associated with the second user to the general-purpose learning model, and outputs an estimated value for each of the one or more pieces of behavioral information.

[0045] The extraction unit 134 extracts behavioral information to be displayed on the information terminal of the first user based on the estimated value that the evaluation unit 133 causes to be output to the behavioral learning model and the estimated value that the evaluation unit 133 causes to be output to the general-purpose learning model. As an example, the extraction unit 134 extracts behavioral information in which the difference between the estimated value that the evaluation unit 133 causes to be output to the behavioral learning model and the estimated value that the evaluation unit 133 causes to be output to the general-purpose learning model is equal to or greater than a threshold value.

[0046] The extraction unit 134 extracts behavioral information for displaying on the information terminal of the first user, the behavioral information in which the magnitude relationship between the estimated value output by the evaluation unit 133 to the behavioral learning model and the estimated value output by the evaluation unit 133 to the general-purpose learning model satisfies a predetermined condition. As an example, the behavioral information may be extracted based on the ratio between the estimated value output by the general-purpose learning model and the estimated value output by the behavioral learning model. For example, the estimated value output by the general-purpose learning model is expressed as P 0 , the estimated value output by the behavioral learning model is P L In this case, the determination unit 133 determines that P L / P 0 may be extracted in such a way that the behavior information is equal to or greater than a predetermined threshold value.

[0047] By configuring the information processing device 1 in this way, when the user's behavior has a high affinity with the behavior of a user who belongs to the community, the user can be notified of information related to the community.

[0048] Behavioral information that has a large distance from the behavioral information of the first user is expected to be information that the first user does not know, and such information may be useful to the first user.

[0049] The extraction unit 134 converts each of the first user's behavior information and the second user's behavior information into vectors, calculates the distance between the vector corresponding to the second user's behavior information and the vector corresponding to the first user's behavior information, and extracts behavior information whose calculated distance is equal to or greater than a predetermined threshold as behavior information to be displayed on the information terminal of the first user. The extraction unit 134 extracts behavior information whose distance from the first user's behavior information is equal to or greater than a predetermined threshold from the second user's behavior information whose estimated value is equal to or greater than a predetermined threshold as behavior information to be displayed on the information terminal of the first user.

[0050] The identification unit 132 may be configured to identify the second user based on the amount of activity in the community. The calculation unit 136 calculates the amount of activity of each of the multiple users in the community based on the behavioral information of the multiple users participating in a specific community stored in the storage unit 12. As an example of the amount of activity, it is the number of comments posted to the community or the amount of behavioral information posted by the user. The calculation unit refers to the storage unit 12, tabulates the behavioral information of each user belonging to the community, and calculates the amount of activity of the user.

[0051] The identification unit 132 identifies a second user from among users whose activity amount calculated by the calculation unit 136 is equal to or greater than a predetermined threshold. As an example, the identification unit 132 identifies a user whose activity amount is equal to or greater than a predetermined threshold as the second user from among users corresponding to behavioral information whose similarity to the behavioral information of the first user is equal to or greater than a predetermined threshold.

[0052] The information to be displayed on the user's information terminal may be determined based on the content of the behavior indicated by the behavior information.

[0053] The determination unit 137 determines information to be displayed on the information terminal of the first user based on the content of the behavior indicated by the behavior information extracted by the extraction unit 134. As an example, the storage unit 12 stores a prediction model, which is a trained model that predicts the probability that the behavior is a behavior for a predetermined product, using behavior information and the product targeted by the behavior information as training data, and is trained to output a product related to the behavior information when the behavior information is input.

[0054] The determination unit 137 inputs the behavioral information extracted by the extraction unit 134 into the prediction model, and outputs the product targeted by the behavioral information. The determination unit 137 identifies a link in the product information that corresponds to the product output by the prediction model, and determines it as information to be displayed on the information terminal 2. The display control unit 135 controls the behavioral information extracted by the extraction unit 134 to be displayed on the information terminal of the first user, and controls the information determined by the determination unit 137 to be associated with the behavioral information extracted by the extraction unit 134 and displayed on the information terminal of the first user.

[0055] The information processing device 1 may be configured to grant a benefit to a second user corresponding to the behavioral information displayed on the information terminal 2 of the first user when the first user purchases a product or the like based on the displayed information.

[0056] The information terminal 2 further includes a benefit granting unit 138 that grants a predetermined benefit to the second user identified by the identification unit 132 when the first user purchases a predetermined product via information for viewing information about the predetermined product that is displayed on the information terminal by the display control unit 135. When displaying a link in the product information that corresponds to a character string indicating a product included in the behavior information, the display control unit 135 causes the information terminal 2 to display a link in which a user ID indicating the second user is embedded.

[0057] When the first user purchases a product on a page showing information about the predetermined product, the acquisition unit 131 acquires information indicating that the first user has purchased the predetermined product. The information indicating that the first user has purchased the predetermined product includes the user ID of the second user embedded in the link. The information indicating that the first user has purchased the predetermined product may include the price of the product purchased by the first user.

[0058] When the benefit granting unit 138 acquires information indicating that the first user has purchased a predetermined product, the benefit granting unit 138 performs processing for granting a predetermined benefit to the second user. As an example, the benefit granting unit 138 transmits an instruction to grant points to the second user to a server that manages points in a predetermined point service. The amount of points granted to the second user may be a predetermined amount, or may be an amount calculated by multiplying the product purchased by the first user by a predetermined coefficient.

[0059] By configuring the information processing device 1 in this way, it is possible to encourage users to actively post to the community.

[0060] [Processing flow in information processing device 1] Fig. 6 is a flowchart showing the flow of processing in the information processing device 1. The flowchart shown in Fig. 6 starts from the point in time when behavioral information is acquired.

[0061] The acquiring unit 131 acquires behavioral information of a first user (S01). The identifying unit 132 identifies a second user who exhibits behavior similar to the behavioral information of the first user (S02).

[0062] The evaluation unit 133 inputs the behavioral information of the second user into the behavior learning model and outputs an estimated value (S03). The extraction unit 134 extracts behavioral information to be displayed on the information terminal of the first user based on the estimated value (S04).

[0063] The display control unit 135 determines whether the extracted behavioral information is a behavior related to a predetermined product (S05). As an example, the display control unit 135 determines whether the extracted behavioral information includes a character string indicating the predetermined product.

[0064] If the extracted behavioral information is a behavior related to a specific product (YES in S05), the display control unit 135 identifies a link corresponding to the product included in the behavioral information, and causes the information terminal 2 to display a screen including the identified link and the extracted behavioral information (S06). Then, the information processing device 1 ends the process.

[0065] If the extracted behavioral information is not a behavior related to a predetermined product (NO in S05), the display control unit 135 causes the information terminal 2 to display a screen including the extracted behavioral information (S07). Then, the information processing device 1 ends the process.

[0066] [Effects of information processing device 1] By configuring the information processing device 1 in this way, it is possible to improve the effectiveness of advertisements on the community site.

[0067] Furthermore, this invention will make it possible to contribute to Goal 9 of the United Nations' Sustainable Development Goals (SDGs), which is to "build resilient infrastructure, promote industry, innovation and infrastructure."

[0068] Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist of the present invention. For example, all or part of the device can be configured by distributing or integrating functionally or physically in any unit. In addition, new embodiments resulting from any combination of multiple embodiments are also included in the embodiments of the present invention. The effect of the new embodiment resulting from the combination combines the effect of the original embodiment. [Explanation of symbols]

[0069] 1. Information processing device 2. Information terminal 11 Communications Department 12 Storage section 13 Control section 131 Acquisition Department 132 Specific part 133 Evaluation Department 134 Extraction part 135 Display control unit 136 Calculation Unit 137 Decision Section 138 Benefits Granting Department

Claims

1. (1) A memory unit that stores behavioral information indicating user behavior, the behavioral information being for each of a plurality of users participating in a specified community, and (2) a behavioral learning model that has learned the behavioral information of each of a plurality of users participating in the specified community, the behavioral learning model being trained to take as input behavioral information indicating user behavior and output an estimated value indicating the degree to which the user behavior indicated by the input behavioral information is estimated to be the behavior of a user belonging to the community; an acquisition unit that acquires behavior information indicating a behavior of a first user who participates in the community; an evaluation unit that, when the acquisition unit acquires the behavioral information of the first user, inputs behavioral information of the second user, which is different from the first user and is identified as a user exhibiting behavior similar to the behavior of the first user based on the behavioral information of the first user, into the behavior learning model and outputs an estimated value; an extraction unit that extracts behavioral information of the second user for determining information to be displayed on an information terminal of the first user based on the estimated value output by the evaluation unit; a determination unit that determines information to be displayed on the information terminal of the first user based on the content of the behavioral information extracted by the extraction unit; a display control unit that controls the information determined by the determination unit to be displayed on an information terminal of the first user; An information processing device having the above configuration.

2. The storage unit further stores a general-purpose learning model that is a pre-trained general-purpose language model, the general-purpose learning model being configured to receive behavioral information as an input and output an estimated value; The behavioral learning model is a trained model obtained by fine-tuning the general-purpose learning model based on behavioral information of a plurality of users participating in the specified community, The evaluation unit inputs the behavioral information of the second user into the general-purpose learning model and further outputs an estimated value; The extraction unit extracts behavioral information to be displayed on the information terminal of the first user based on the estimated value that the evaluation unit causes to be output to the behavioral learning model and the estimated value that the evaluation unit causes to be output to the general-purpose learning model. The information processing device according to claim 1 .

3. The extraction unit extracts behavioral information in which a magnitude relationship between the estimated value that the evaluation unit causes to be output to the behavioral learning model and the estimated value that the evaluation unit causes to be output to the general-purpose learning model satisfies a predetermined condition, and the behavioral information is to be displayed on the information terminal of the first user. The information processing device according to claim 2 .

4. the extraction unit converts each of the first user's behavior information and the second user's behavior information into a vector, calculates a distance between the vector corresponding to the second user's behavior information and the vector corresponding to the first user's behavior information, and extracts behavior information for which the calculated distance is equal to or greater than a predetermined threshold as behavior information to be displayed on the information terminal of the first user. The information processing device according to claim 1 .

5. the display control unit controls the behavioral information extracted by the extraction unit to be displayed on the information terminal of the first user, and controls the information determined by the determination unit to be associated with the behavioral information extracted by the extraction unit and displayed on the information terminal of the first user. The information processing device according to claim 1 .

6. The behavioral information is text data indicating user comments in the community. the display control unit controls the information terminal of the first user to further display the behavioral information extracted by the extraction unit, and when the behavioral information extracted by the extraction unit includes a character string indicating a predetermined product, to display information for viewing information about the predetermined product on the information terminal of the first user in association with the behavioral information extracted by the extraction unit. The information processing device according to claim 5 .

7. the display control unit further displays the behavioral information extracted by the extraction unit on the information terminal of the first user, and when the behavioral information extracted by the extraction unit indicates a behavior related to a predetermined product, displays information for viewing information related to the predetermined product on the information terminal of the first user in association with the behavioral information extracted by the extraction unit; The information processing device further includes a benefit granting unit that grants a predetermined benefit to the second user when the first user purchases the predetermined product via information for viewing information about the predetermined product displayed on the information terminal by the display control unit. The information processing device according to claim 1 .

8. The computer executes An acquisition step of acquiring behavior information indicating a behavior of a first user who participates in a predetermined community; an evaluation step in which, when the behavioral information of the first user is acquired in the acquisition step, a behavioral learning model that has learned the behavioral information of each of a plurality of users participating in the specified community and stored in a storage unit, the behavioral learning model being trained to use behavioral information as input and output an estimated value indicating the degree to which the behavior of the user indicated by the input behavioral information is estimated to be the behavior of a user belonging to the community, refers to the behavioral information of each of the plurality of users participating in the specified community stored in the storage unit, inputs behavioral information of the second user that is different from the first user and that is identified as a user that exhibits behavior similar to the behavior of the first user based on the behavioral information of the first user, and outputs an estimated value; an extraction step of extracting behavioral information of the second user for determining information to be displayed on an information terminal of the first user based on the estimated value output in the evaluation step; a determination step of determining information to be displayed on the information terminal of the first user based on the content of the behavior information extracted in the extraction step; a display control step of further displaying the information determined in the determination step on an information terminal of the first user; An information processing method comprising the steps of:

9. On the computer, An acquisition step of acquiring behavior information indicating a behavior of a first user who participates in a predetermined community; an evaluation step in which, when the behavioral information of the first user is acquired in the acquisition step, a behavioral learning model that has learned the behavioral information of each of a plurality of users participating in the specified community and stored in a storage unit, the behavioral learning model being trained to use behavioral information as input and output an estimated value indicating the degree to which the behavior of the user indicated by the input behavioral information is estimated to be the behavior of a user belonging to the community, refers to the behavioral information of each of the plurality of users participating in the specified community stored in the storage unit, inputs behavioral information of the second user that is different from the first user and that is identified as a user that exhibits behavior similar to the behavior of the first user based on the behavioral information of the first user, and outputs an estimated value; an extraction step of extracting behavioral information of the second user for determining information to be displayed on an information terminal of the first user based on the estimated value output in the evaluation step; a determination step of determining information to be displayed on the information terminal of the first user based on the content of the behavior information extracted in the extraction step; a display control step of further displaying the information determined in the determination step on an information terminal of the first user; A program for executing the above.

Images