Information processing device, information processing method, and information processing program

Abstract

To provide an information processing device, an information processing method, and an information processing program, capable of more appropriately obtaining discussions related to a topic.SOLUTION: An information processing device comprises an acquisition unit, a determination unit, and a generation unit. The acquisition unit acquires user information, which is information about each user who has performed a specific action in an online service. The determination unit determines a plurality of distinct personas through clustering based on the user information acquired by the acquisition unit. The generation unit causes a generative AI to generate information indicative of a discussion on a specific topic performed by the plurality of personas on the basis of the information indicative of the plurality of personas determined by the determination unit.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and an information processing program.

Background Art

[0002] Conventionally, techniques for automatically generating opinions on issues have been proposed. For example, Patent Document 1 discloses a technique of searching for articles using keywords and issue words obtained by analyzing an input issue, extracting sentences related to the issue from among them, rearranging them to generate and evaluate a sentence, and outputting the sentence with the highest evaluation as an opinion sentence on the issue.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the above conventional technology, since an opinion sentence is generated based on rules, it may be difficult to appropriately generate an opinion sentence on an issue, and there is room for improvement in obtaining a more appropriate discussion on the issue.

[0005] The present application has been made in view of the above, and an object thereof is to provide an information processing apparatus, an information processing method, and an information processing program capable of obtaining a more appropriate discussion on an issue.

Means for Solving the Problems

[0006] The information processing apparatus according to the present application includes an acquisition unit, a determination unit, and a generation unit. The acquisition unit acquires user information, which is information of each user who has performed a specific action in an online service. The determination unit determines a plurality of different personas by clustering based on the user information acquired by the acquisition unit. The generation unit causes an AI that generates information indicating a discussion on a specific topic by the plurality of personas to generate information indicating a discussion on a specific topic based on the information indicating the plurality of personas determined by the determination unit.

Effect of the Invention

[0007] According to one aspect of the embodiment, a more appropriate discussion on the topic can be obtained.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Embodiment for Carrying Out the Invention

[0009] Hereinafter, embodiments for implementing the information processing apparatus, information processing method, and information processing program according to the present application (hereinafter referred to as "embodiments") will be described in detail with reference to the drawings. Note that the information processing apparatus, information processing method, and information processing program according to the present application are not limited by this embodiment. Also, the respective embodiments can be appropriately combined as long as the processing contents do not conflict. In addition, in the following embodiments, the same parts are denoted by the same reference numerals, and redundant explanations are omitted.

[0010] 〔1. Example of Information Processing〕 First, an example of information processing according to the embodiment will be described with reference to FIG. 1. FIG. 1 is a diagram showing an example of information processing according to the embodiment.

[0011] The information processing apparatus 1 shown in FIG. 1 is an information processing apparatus for realizing the generation of information indicating a discussion on a specific topic, and provides a discussion information providing service for providing information indicating a discussion on a specific topic. The information processing apparatus 1 is realized by, for example, one or more servers or a cloud system.

[0012] As shown in FIG. 1, the information processing apparatus 1 receives a discussion information request transmitted from the terminal device 3 of the user M who uses the discussion information providing service (step S1). The discussion information request includes, for example, acquisition target information, topic information, end condition information, and constraint condition information.

[0013] The acquisition target information is information for designating information used for determining a persona in step S3 described later. For example, the acquisition target information includes type information indicating the type of online service designated by the user M and target action specifying information for specifying a target action that is the action designated by the user M in the online service. The target action is an example of a specific action.

[0014] The types of online games include, for example, online Q&A (Question and Answer) services, online search services, online news provision services, Internet bulletin board services, etc., which are online services provided by the information processing apparatus 4, but are not limited to such examples.

[0015] The target action is, for example, the use of services in a specific category if it is an online Q&A service, an Internet bulletin board service, or an online news provision service, or a search using a specific search query if it is an online search service, but is not limited to such examples.

[0016] The use of services in a specific category is, for example, the viewing of information (e.g., information indicating questions or answers) in a specific category if it is an online Q&A service, but is not limited to such examples. For example, the use of services in a specific category may be, instead of viewing information in a specific category, the posting of information (e.g., information indicating questions or answers) in a specific category.

[0017] A search using a specific search query is, for example, a search in an online search service using a search query including a specific search keyword or a specific phrase, but is not limited to such examples. For example, a search using a specific search query may be a search in an online Q&A service using a search query including a specific search keyword or a specific phrase.

[0018] The target action specific information is information indicating specific terms related to an online Q&A service, and is, for example, information indicating one or more specific terms included in one or more categories. In this case, the use of services in the category including the specific terms is the target action. In an online Q&A service, categories are represented in a hierarchical structure.

[0019] For example, the category "childbirth" is expressed as the top layer "child-rearing and school", the middle layer "child-rearing", and the bottom layer "childbirth", etc., but is not limited to such examples, and may be expressed in 4 or more layers, or may be expressed in 2 or fewer layers.

[0020] Specific terms are, for example, in the case of categories related to child-rearing, "elementary school", "pregnancy", "childbirth", "early childhood education", "kindergarten", "nursery school", "exams", "child", "kids", "high school", "family", etc., but are not limited to such examples.

[0021] Also, the target action specific information may be, for example, information on the category name, or may be information on the category shown in a hierarchical structure such as, for example, the information of the character string "child-rearing and school > child-rearing > childbirth", etc., but is not limited to such examples. For example, the information for specifying the category indicated by the target action specific information may be information on the category shown in the top layer, the middle layer, or the bottom layer.

[0022] Also, the target action specific information is, for example, information for specifying a specific search query in a search service, and includes, for example, information such as a specific search keyword or a specific search phrase. In this case, the search using the search query that includes a part or all of the specific search keyword or the specific search phrase is the target action.

[0023] The issue information is information indicating a specific issue designated by the user M. The end condition information is information indicating the end condition of the discussion regarding the specific issue. The constraint condition information is information indicating the constraint conditions of the discussion by a plurality of personas determined in step S3 described later.

[0024] For example, assume that user M is an employee of a travel agency. In this case, the topic is, for example, "Travel policies for the child-rearing generation." Also, the end condition is, for example, "Based on everyone's opinions, summarize one specific travel policy for the child-rearing generation." Also, the constraint condition is, for example, "Many in the child-rearing generation stay for 1 night when traveling, and we want to implement policies to extend this to 2 nights or more. We hope that policies are considered based on the anxieties and needs of the child-rearing generation regarding travel."

[0025] Also, assume that user M is an employee of a developer. In this case, the topic is, for example, "Housing (custom-built houses, detached houses for sale, condominiums, etc.) that the child-rearing generation would like to live in." Also, the end condition is, for example, "Based on everyone's opinions, decide on three child-rearing support plans to create a town where people would be more willing to raise children." Also, the constraint condition is, for example, "We want to know what policies the child-rearing generation is seeking in order to overcome the low birthrate and aging population."

[0026] Also, assume that user M is a local government employee. In this case, the topic is, for example, "Child-rearing support policies." Also, the end condition is, for example, "Summarize three conditions for housing that the child-rearing generation would like to live in." Also, the constraint condition is, for example, "As a developer, we want to understand the housing requirements of the child-rearing generation."

[0027] User M can operate the terminal device 3 to input or select acquisition target information, topic information, end condition information, constraint condition information, etc. The terminal device 3 transmits a discussion information request including acquisition target information, topic information, end condition information, constraint condition information, etc. input or selected by user M to the information processing device 1.

[0028] Subsequently, the information processing device 1 acquires user information, which is information of each user U who has performed the target action specified by the acquisition target information, from the information processing device 4 or the internal storage unit, etc., of the online service of the type specified by the acquisition target information included in the discussion information request received in step S1 (step S2).

[0029] The user information includes the attribute information and behavior information of user U. The attribute information of user U is, for example, information indicating the attributes of user U. The attributes of user U are, for example, the demographic attributes of user U, but may also be the psychographic attributes of user U, or a combination of the demographic attributes and psychographic attributes of user U.

[0030] The behavior information of user U is information regarding the target behavior performed by user U for the target behavior specified by the information to be acquired. For example, when the type of online service specified by the information to be acquired is an online Q&A service, the behavior information of user U is information indicating the usage content of the service in a specific category that is the category specified by the target behavior specifying information.

[0031] The usage content of the service in a specific category is, for example, browsing of information in a specific category (such as information indicating a question or an answer), posting of information (such as information indicating a question or an answer) to a specific category, etc., but is not limited to such examples.

[0032] Also, when the type of online service specified by the information to be acquired is an online search service, the behavior information of user U is, for example, information indicating the content browsed by performing a search with a search query including a specific search keyword or a specific phrase indicated by the target behavior specifying information, but is not limited to such examples.

[0033] Subsequently, the information processing apparatus 1 determines a plurality of different personas by clustering based on the user information of each user U acquired in step S1 (step S3).

[0034] For example, assume that the type of online service indicated by the type information of the information to be acquired included in the discussion information request is an online Q&A service, and the specific behavior indicated by the target specifying information of the information to be acquired included in the discussion information request is the usage of the service in a category including a specific term.

[0035] In this case, the information processing apparatus 1 extracts a plurality of characteristic words from the information of a specific category, which is a category including a specific term, performs topic classification for each specific category using LDA (Latent Dirichlet Allocation) from these plurality of characteristic words, and clusters the user U.

[0036] For example, the information processing apparatus 1 extracts a plurality of characteristic words for each specific category from the information of a specific category, which is a category including a specific term indicated by the target behavior specific information. The information processing apparatus 1 extracts, for example, individual words included in the information of a specific category, and removes words having no meaning from the extracted plurality of words.

[0037] The information processing apparatus 1 acquires characteristic words as vectors using TF-IDF (Term Frequency-Inverse Document Frequency) or Bag-of-Words or the like from the plurality of words from which words having no meaning have been removed. The information processing apparatus 1 uses LDA to determine the topic of a specific category for each specific category from the plurality of characteristic words extracted as vectors, and performs topic classification for classifying each specific category into the corresponding topic. The information processing apparatus 1 clusters the user U by classifying the user U using a specific category into the topic corresponding to that specific category. Note that the clustering by the information processing apparatus 1 is not limited to that by LDA, and k-means or any other arbitrary clustering method can be used.

[0038] The user U using a specific category classified into each of a plurality of different topics is classified into each of the plurality of topics. For example, the user U can also be classified into the topic with the largest number of uses of a specific category.

[0039] The information processing apparatus 1 extracts, as target topics, topics in which the ratio of the user U is equal to or higher than a threshold based on the ratio of the user U classified into each topic, and determines a persona corresponding to such target topics. The threshold is 5%, but is not limited to such an example.

[0040] The information processing apparatus 1 determines a persona corresponding to a target topic based on, for example, the attribute information of the user U classified into the target topic and the target topic information that is information on the target topic.

[0041] For example, the information processing apparatus 1 determines, as information on the persona corresponding to the target topic, information including an attribute with a high ratio among the attributes of the user U classified into the target topic and information on a specific category classified into the target topic. The attributes included in the information on the persona are indicated by, for example, gender and age, but are not limited to such examples. For example, in addition to or instead of gender and age, it may be indicated by family composition, annual income, place of residence, occupation, etc.

[0042] Also, for example, the information processing apparatus 1 rounds the ones digit of the ratio of the users U belonging to a topic with a ratio of the user U of a threshold value or more, and then divides the value by 10 to set the resulting number as the number of personas corresponding to the topic, but is not limited to such an example.

[0043] When the number of personas corresponding to the same target topic is one, the information processing apparatus 1 sets the attribute with the highest ratio among the attributes of the user U classified into the target topic as the attribute of the persona. Also, when the number of personas corresponding to the same target topic is two or more, the information processing apparatus 1 sets each of two or more attributes in the order of the attributes with a high ratio among the attributes of the user U classified into the target topic as the attribute of the corresponding persona among the two or more personas.

[0044] The target topic information included in the information on the persona is information indicating the target topic, but may be information obtained by summarizing information on a specific category classified into the target topic using a text generation AI (Artificial Intelligence) or the like.

[0045] A text generation AI is, for example, a language model trained to estimate and output the next token from an input token sequence, such as a transformer-based model or an RNN (Recurrent Neural Network)-based model.

[0046] Transformer-based models include, for example, GPT (Generative Pre-trained Transformer) and BARD (Bidirectional Auto Regressive Dialogues), but are not limited to such examples. RNN-based models include, for example, RWKV (Receptance Weighted Key Value), but are not limited to such examples. Note that it is desirable to perform learning so that the input information is not used as a new answer, thereby concealing information such as personal information in the input.

[0047] For example, in step S3, assume that the topics to be classified for user U are topics 1 to 10, and the target topics are topic 2, topic 4, topic 5, and topics 7 to 10. Also assume that the number of personas corresponding to topic 4, topic 9, and topic 10 is 2.

[0048] In this case, the information processing apparatus 1 determines, for example, 10 personas such as a 30-year-old woman with topic 2 as the background, a 20-year-old woman with topic 4 as the background, a 30-year-old woman with topic 4 as the background, a 20-year-old woman with topic 5 as the background, a 30-year-old woman with topic 7 as the background, a 30-year-old woman with topic 8 as the background, a 30-year-old woman with topic 9 as the background, a 40-year-old woman with topic 9 as the background, a 20-year-old woman with topic 10 as the background, and a 30-year-old woman with topic 10 as the background.

[0049] Subsequently, the information processing apparatus 1 causes the generation AI to generate information indicating a discussion on a specific issue by these multiple personas based on the information indicating the multiple personas determined in step S3 (step S4). The generation AI is the text generation AI or the multimodal AI described above.

[0050] The multimodal AI is, for example, a model that generates an image from text or generates text from an image, such as GPT-4-Trubo, GPT-4V, CM3Leon (Chameleon Multimodal Model), etc., but is not limited to such examples.

[0051] In step S4, the information processing apparatus 1 inputs, as input information to the generation AI, information including instruction information instructing opinions on a specific issue by the multiple personas determined in step S3, and causes the generation AI to generate information indicating a discussion by the multiple personas. The specific issue is the issue indicated by the issue information included in the discussion information request.

[0052] The instruction information includes, for example, issue information indicating a specific issue, persona information including information of each of the multiple personas, the past speech history of the personas, and output instruction information instructing the selection of the persona to speak from among the multiple personas indicated by the persona information and the output of the speech to the previously selected persona.

[0053] The information processing apparatus 1 repeats the process of causing the generation AI to output output information including information indicating the persona selected as the next speaker and information indicating the speech of the persona selected as the current speaker by inputting the information including the instruction information to the generation AI as input information. Thereby, the information processing apparatus 1 can cause the generation AI to generate information indicating a discussion by the multiple personas.

[0054] In addition, when the generative AI is GPT-4 of OpenAI, information such as topic information indicating a specific topic, persona information including information on each of a plurality of personas, and output instruction information is set as information in the system message, and information indicating the selected persona and the past speech history of the persona are set as the user prompt, but it is not limited to such an example.

[0055] Also, the instruction information includes, for example, information such as the string "Discussions will be held among the following 10 users. The details of the topic and the 10 users are as follows." Thus, information indicating discussions by a plurality of personas can be appropriately generated by the generative AI. The detailed information of the user is persona information and includes topic information and attribute information as described above. Note that the persona information may include information on categories with high usage frequency by user U instead of topic information, for example.

[0056] In addition, the information processing apparatus 1 can also input, as input information to the generative AI, information including the end condition information included in the discussion information request in the instruction information. The end condition information is information indicating the end condition of the discussion regarding a specific topic.

[0057] In this case, the information processing apparatus 1 includes, for example, information such as the string "Discussions will be held among the following 10 users. The topic, the end condition of the discussion, and the detailed information of the 10 users are as follows." in the instruction information, together with the topic information, the end condition information, information indicating a plurality of personas, and the output instruction information.

[0058] In addition, the information processing apparatus 1 can also input, as input information to the generative AI, information including the constraint condition information included in the discussion information request in the instruction information. The constraint condition information is information indicating the constraint conditions for discussions by a plurality of personas and is information indicating the user M's request for the content of the discussion.

[0059] In this case, the information processing apparatus 1 includes, for example, the information of the character string "Discussions will be held among the following 10 users. The topic, the end condition of the discussion, and the detailed information of the 10 users are as follows." and the topic information, the end condition information, the information indicating a plurality of personas, and the output instruction information, etc. in the instruction information.

[0060] Further, the information processing apparatus 1 can also input, as input information, the information including the information instructing the progress of the discussion by the facilitator who is the progress manager for summarizing the opinions by a plurality of personas into the instruction information and input it to the generation AI.

[0061] In this case, the information processing apparatus 1 includes, for example, the role information which is the information indicating the role of the facilitator in the instruction information. The role information includes, for example, the information of the character string "#Facilitator\nYou are the progress manager for summarizing the conversation. The facilitator should assign turns to speak to the 10 users. After listening to all the users' opinions and when the conversation is summarized, the facilitator should summarize the conversation again and assign turns to speak to the users again to dig deeper into the specific content.", but is not limited to such an example.

[0062] When the information processing apparatus 1 instructs the progress of the discussion by the facilitator, the instruction information includes the information instructing the facilitator to speak first. Thereby, the information processing apparatus 1 can appropriately generate, by the generation AI, the information indicating the discussion by a plurality of personas.

[0063] Note that the above-described generation AI is arranged in an external information processing apparatus, and the information processing apparatus 1 obtains the information indicating the discussion by generating it by the generation AI via the API (Application Programming Interface) provided by the external information processing apparatus, but is not limited to such an example. For example, the generation AI may be arranged in the information processing apparatus 1.

[0064] Subsequently, the information processing apparatus 1 provides the discussion information to the user M by transmitting the discussion information to the terminal device 3 of the user M (step S5). The user M is an example of the target person to whom the discussion information is provided.

[0065] For example, every time there is a statement from a persona or a facilitator, the information processing device 1 can provide the discussion information to the user M. As a result, the user M can grasp the progress of the discussion in real time.

[0066] The user M can operate the terminal device 3 to input a question regarding the discussion into the terminal device 3. When the terminal device 3 receives an input of a question regarding the discussion from the user M, the terminal device 3 transmits question information, which is information indicating the question regarding the discussion, to the information processing device 1.

[0067] The information processing device 1 receives the question information transmitted from the terminal device 3 (step S6). When the information processing device 1 receives the question information, the information processing device 1 inputs the information including the question information into the generation AI as input information further including instruction information.

[0068] For example, assume that the topic is "Housing where child-rearing generations want to live (custom-built houses, detached houses for sale, condominiums, etc.)". The user M can input, as question information, information such as the character string "Human: I've been listening to everyone's opinions and I've come to the conclusion that the surrounding environment is more important than the layout and functions of the house. Is there any mistake in this recognition?" while looking at the discussion of the persona.

[0069] In this case, the information processing device 1 generates instruction information including the information of the character string "Human: I've been listening to everyone's opinions and I've come to the conclusion that the surrounding environment is more important than the layout and functions of the house. Is there any mistake in this recognition?" in the past speech history of the persona. Thereby, the information processing device 1 can involve the user M in the discussion and make the discussion by a plurality of personas proceed more appropriately.

[0070] In addition, the user M can input additional persona information, which is information indicating an additional persona that is not one of the plurality of personas, into the terminal device 3 by operating the terminal device 3. When the terminal device 3 receives the input of the additional persona information from the user M, the terminal device 3 transmits question information, which is the additional persona information, to the information processing device 1. The additional persona information includes, for example, attribute information of the additional persona and information on topics corresponding to the additional persona.

[0071] The information processing device 1 receives the additional persona information transmitted from the terminal device 3 (step S7). When the information processing device 1 receives the additional persona information, the information processing device 1 inputs, as input information, information including the persona information with the additional persona information added into the generative AI.

[0072] Thereby, the information processing device 1 can input, as input information, information indicating a discussion by the additional persona indicated by the additional persona information and the above-described plurality of personas into the generative AI as instruction information.

[0073] In addition, when the discussion ends, the information processing device 1 creates a mind map of the discussion on a specific topic by the plurality of personas based on the information indicating the discussion on the specific topic by the plurality of personas (step S8).

[0074] The information processing device 1 inputs, as input information, information including, for example, the information of the character string "Input the conversation of a certain discussion. Please summarize the conversation in a mind map. The output should be text data in a format corresponding to the syntax of mermaid.js." and the discussion information into the generative AI, and causes the generative AI to output data used for generating the mind map.

[0075] Then, the information processing apparatus 1 creates a mind map based on the text data output from the generative AI. Note that when the generative AI is a multimodal AI or an image generation AI, the information processing apparatus 1 can also directly cause the generative AI to create a mind map. The image generation AI is, for example, StackGAN (Generative Adversarial Networks), AttnGAN, T2I (Text-to-Image) with Transformers, DALL-E, etc., but is not limited to such examples.

[0076] In this way, the information processing apparatus 1 determines a plurality of different personas by clustering based on the user information, which is the information of each user U who has performed a specific action in the online service, and causes the generative AI to generate information indicating a discussion on a specific topic by the plurality of personas based on the information indicating the plurality of personas. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the topic.

[0077] Hereinafter, the configuration of the information processing system including the information processing apparatus 1, the plurality of terminal devices 2, the terminal device 3, and the information processing apparatus 4 that perform such processing will be described in detail.

[0078] 〔2. Configuration of Information Processing System〕 FIG. 2 is a diagram showing an example of the configuration of the information processing system according to the embodiment. As shown in FIG. 2, the information processing system 100 according to the embodiment includes an information processing apparatus 1, a plurality of terminal devices 2, a terminal device 3, and an information processing apparatus 4.

[0079] The plurality of terminal devices 2 are used by different users U. The terminal device 3 is, for example, a terminal device of a user M such as an employee of a company or a staff member of a local government. The terminal devices 2 and 3 are, for example, a notebook PC (Personal Computer), a desktop PC, a smartphone, a tablet PC, or a wearable device. The wearable device is, for example, smart glasses or a smart watch, etc., but is not limited to such examples.

[0080] The information processing apparatus 4 provides various online services to the user U. For example, the information processing apparatus 4 provides an online Q&A service, an online search service, an online news providing service, a net bulletin board service, etc. to the user U, but is not limited to such examples.

[0081] Each of the information processing apparatus 1, the terminal apparatus 2, the terminal apparatus 3, and the information processing apparatus 4 is connected to be communicable with each other by wire or wirelessly via the network N. Note that the information processing system 100 shown in FIG. 2 may include a plurality of information processing apparatuses 1 and the like.

[0082] The network N includes, for example, a WAN (Wide Area Network) such as the Internet and a mobile communication network such as LTE (Long Term Evolution), 4G (4th Generation), and 5G (5th Generation: the 5th generation mobile communication system).

[0083] The terminal apparatuses 2 and 3 are connected to the network N via a mobile communication network, short-range wireless communication such as Bluetooth (registered trademark), or a wireless LAN (Local Area Network), and can communicate with the information processing apparatus 1, the information processing apparatus 4, and the like.

[0084] 〔3. Configuration of Information Processing Apparatus 1〕 FIG. 3 is a diagram showing an example of the configuration of the information processing apparatus 1 according to the embodiment. As shown in FIG. 3, the information processing apparatus 1 includes a communication unit 10, a storage unit 11, and a processing unit 12.

[0085] 〔3.1. Communication Unit 10〕 The communication unit 10 is realized by, for example, a communication module or a NIC (Network Interface Card). Then, the communication unit 10 is connected to the network N by wire or wirelessly, and transmits and receives information to and from various other devices. For example, the communication unit 10 transmits and receives information to and from each of the terminal apparatus 2, the terminal apparatus 3, and the information processing apparatus 4 via the network N.

[0086] [3.2. Memory Unit 11] The memory unit 11 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The memory unit 11 has a user information storage unit 20.

[0087] [3.2.1. User Information Storage Unit 20] The user information storage unit 20 stores user information including information about the user U. FIG. 4 is a diagram showing an example of a user information table stored in the user information storage unit 20 of the information processing apparatus 1 according to the embodiment. As shown in FIG. 4, the user information table stored in the user information storage unit 20 includes items such as "user ID", "attribute information", and "behavior history".

[0088] The "user ID" is identification information for identifying the user U. The "attribute information" is attribute information of the user U corresponding to the "user ID", and includes, for example, information on psychographic attributes or demographic attributes. Demographic attributes are, for example, gender, age, place of residence, and occupation, and psychographic attributes are objects of interest such as travel, clothing, cars, and religion, lifestyle, thoughts, and trends of thoughts.

[0089] The "behavior history" is the behavior history of the user U in the online service, and includes information such as a search history, a browsing history, a posting history, and a purchase history. The search history is information on search queries used in the past by the user U and content browsed by the user U from among the search results. The information on the search query is, for example, information such as a search keyword or a search phrase.

[0090] The browsing history includes information indicating the content viewed by user U in, for example, an online service, and the posting history includes information indicating the content (such as reviews and comments) posted by user U in the past in, for example, an online service. The purchase history includes information on the transaction targets with which user U has conducted transactions in the past.

[0091] [3.3. Processing Unit 12] The processing unit 12 is a controller and is realized, for example, by various programs (corresponding to an example of an information processing program) stored in the storage device inside the information processing apparatus 1 being executed with a RAM or the like as a work area by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit).

[0092] Also, the processing unit 12 is a controller and may be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a GPGPU (General Purpose Graphic Processing Unit).

[0093] As shown in FIG. 3, the processing unit 12 includes a reception unit 30, an acquisition unit 31, a determination unit 32, a generation unit 33, a creation unit 34, and a provision unit 35, and realizes or executes the functions and operations of information processing described below. Note that the internal configuration of the processing unit 12 is not limited to the configuration shown in FIG. 3, and other configurations may be used as long as they can perform the information processing described later.

[0094] [3.3.1. Reception Unit 30] The reception unit 30 receives various requests and information via the network N and the communication unit 10. For example, the reception unit 30 receives a discussion information request transmitted from the terminal device 3. The discussion information request includes, for example, acquisition target information, topic information, end condition information, and constraint condition information.

[0095] The acquisition target information is information that specifies the information used for persona determination. For example, it includes type information indicating the type of online service specified by user M, and target action specification information that specifies the target action, which is the action specified by user M in the online service. The target action is an example of a specific action.

[0096] Examples of the type of online game include, but are not limited to, an online Q&A (Question and Answer) service, an online search service, an online news providing service, a net bulletin board service, etc., which are online services provided by the information processing apparatus 4.

[0097] For example, if the target action is a service such as an online Q&A service, a net bulletin board service, or an online news providing service, it is the use of a service in a specific category. If it is an online search service, it is a search using a specific search query, but is not limited to such examples.

[0098] For example, if it is an online Q&A service, the use of a service in a specific category is the viewing of information (e.g., information indicating a question or an answer) in a specific category, but is not limited to such examples. For example, the use of a service in a specific category may be, instead of viewing information in a specific category, the posting of information (e.g., information indicating a question or an answer) in a specific category.

[0099] A search using a specific search query is, for example, a search in an online search service using a search query that includes a specific search keyword or a specific phrase, but is not limited to such examples. For example, a search using a specific search query may be a search in an online Q&A service using a search query that includes a specific search keyword or a specific phrase.

[0100] The target action specific information is information indicating specific terms related to the online Q&A service, for example, information indicating one or more specific terms included in one or more categories. In this case, the use of the service in the category containing the specific term is the target action. In the online Q&A service, the category is represented in a hierarchical structure.

[0101] For example, the category "childbirth" is represented as the top layer "child rearing and school", the middle layer "child rearing", and the bottom layer "childbirth", etc., but is not limited to such examples and may be represented by 4 or more layers or 2 or fewer layers.

[0102] Specific terms are, for example, in the case of categories related to child rearing, "elementary school", "pregnancy", "childbirth", "early childhood education", "kindergarten", "nursery school", "examination", "child", "kids", "high school", "family", etc., but are not limited to such examples.

[0103] Also, the target action specific information may be, for example, information of the category name, for example, information of the category shown in a hierarchical structure such as the information of the character string "child rearing and school > child rearing > childbirth", etc., but is not limited to such examples. For example, the information specifying the category indicated by the target action specific information may be information of the category shown in the top layer, the middle layer, or the bottom layer.

[0104] Also, the target action specific information is, for example, information specifying a specific search query in the search service, and includes, for example, information such as a specific search keyword or a specific search phrase. In this case, the search with the search query that partially or entirely includes the specific search keyword or the specific search phrase is the target action.

[0105] The issue information is information indicating a specific issue specified by the user M. The end condition information is information indicating the end condition of the discussion regarding the specific issue. The constraint condition information is information indicating the constraint conditions of the discussion by a plurality of personas determined by the determination unit 32.

[0106] For example, assume that user M is an employee of a travel agency. In this case, the issue is, for example, "Travel policies for the child-raising generation". Also, the end condition is, for example, "Based on everyone's opinions, summarize one specific travel policy for the child-raising generation." Also, the constraint conditions are, for example, "Many in the child-raising generation stay for one night when traveling, and we want to implement measures to extend this to two nights or more. We hope that the measures are considered based on the anxieties and needs of the child-raising generation regarding travel."

[0107] Also, assume that user M is an employee of a developer. In this case, the issue is, for example, "Housing (custom-built houses, detached houses for sale, condominiums, etc.) that the child-raising generation would like to live in." Also, the end condition is, for example, "Based on everyone's opinions, please decide on three child-rearing support plans to create a town where one would be more willing to raise children." Also, the constraint conditions are, for example, "We want to know what measures the child-raising generation is seeking in order to overcome the low birthrate and aging population."

[0108] Also, assume that user M is a local government official. In this case, the issue is, for example, "Child-rearing support measures." Also, the end condition is, for example, "Summarize the three conditions for housing that the child-raising generation would like to live in." Also, the constraint conditions are, for example, "As a developer, we want to understand the housing requirements of the child-raising generation."

[0109] Also, the reception unit 30 receives the question information transmitted from the terminal device 3. User M operates the terminal device 3 to input a question regarding the discussion into the terminal device 3. When the terminal device 3 receives the input of a question regarding the discussion from user M, it transmits question information, which is information indicating the question regarding the discussion, to the information processing device 1.

[0110] Also, the reception unit 30 receives the additional persona information transmitted from the terminal device 3. The additional persona information includes, for example, the attribute information of the additional persona and the information of the topic corresponding to the additional persona.

[0111] By operating the terminal device 3, the user M can input additional persona information, which is information indicating an additional persona that is not one of the plurality of personas, into the terminal device 3. When the terminal device 3 receives the input of the additional persona information from the user M, the terminal device 3 transmits question information, which is the additional persona information, to the information processing device 1.

[0112] [3.3.2. Acquisition Unit 31] The acquisition unit 31 acquires various information via the network N and the communication unit 10. For example, the acquisition unit 31 acquires various information from the terminal device 2, the terminal device 3, the information processing device 4, and the like.

[0113] The acquisition unit 31 acquires user information, which is information of each user U who has performed a specific action in the online service, from the information processing device 4, the storage unit 11, or the like. For example, when a discussion information request is received by the reception unit 30, the acquisition unit 31 acquires user information, which is information of each user U who has performed the target action specified by the acquisition target information in the online service of the type specified by the acquisition target information included in the discussion information request, from the information processing device 4, the storage unit 11, or the like.

[0114] The user information includes the attribute information and the action information of the user U. The attribute information of the user U is, for example, information indicating the attribute of the user U. The attribute of the user U is, for example, the demographic attribute of the user U, but may be the psychographic attribute of the user U, or may be a combination of the demographic attribute and the psychographic attribute of the user U.

[0115] The action information of the user U is information regarding the target action performed by the user U who has performed the target action specified by the acquisition target information. The target action is an example of a specific action. For example, when the type of the online service specified by the acquisition target information is an online Q&A service, the action information of the user U is information indicating the usage content of the service in a specific category, which is the category specified by the target action specification information.

[0116] The usage content of a specific category is, for example, browsing information of a specific category (such as information indicating a question or information indicating an answer), posting information to a specific category (such as information indicating a question or information indicating an answer), etc., but is not limited to such examples.

[0117] Also, when the type of the online service specified by the information to be acquired is an online search service, the behavior information of the user U is information indicating the content browsed by performing a search with a search query including a specific search keyword or a specific phrase indicated by the target behavior specifying information, etc., but is not limited to such examples.

[0118] For example, the behavior information of the user U may be the browsing history of the past content of the user U that performed a search with a search query including a specific search keyword or a specific phrase indicated by the target behavior specifying information.

[0119] [3.3.3. Decision Unit 32] The decision unit 32 makes various decisions. For example, the decision unit 32 determines a plurality of different personas by clustering based on the user information acquired by the acquisition unit 31.

[0120] For example, assume that the type of the online service indicated by the type information of the information to be acquired included in the discussion information request is an online Q&A service, and the specific behavior indicated by the target specifying information of the information to be acquired included in the discussion information request is a category including a specific term.

[0121] In this case, the decision unit 32, for example, extracts a plurality of words from the information of a specific category that is a category including a specific term, performs topic classification of each specific category using LDA from these plurality of words, and clusters the user U.

[0122] For example, the determination unit 32 extracts a plurality of characteristic words for each specific category from the information of a specific category that is a category including a specific term indicated by the target action specific information. For example, the determination unit 32 extracts individual words included in the information of a specific category and removes words having no meaning from the extracted plurality of words.

[0123] The determination unit 32 obtains characteristic words as vectors using, for example, TF-IDF or Bag-of-Words from the plurality of words from which words having no meaning have been removed. The determination unit 32 performs topic classification for each specific category using LDA from the plurality of characteristic words extracted as vectors, and classifies each specific category into the corresponding topic. The determination unit 32 clusters the user U by classifying the user U who uses a specific category into the topic corresponding to that specific category.

[0124] The information of a specific category is information indicating a specific category name, but is not limited to such an example. For example, instead of or in addition to the information indicating a specific category name, information posted in a specific category (for example, information indicating a question or an answer) may be included.

[0125] Also, assume that the type of the online service indicated by the type information of the acquisition target information included in the discussion information request is an online search service, and the specific action indicated by the target specific information of the acquisition target information included in the discussion information request is a search with a specific search query.

[0126] In this case, for example, the determination unit 32 extracts a plurality of words from the content browsed by performing a search with a search query including a specific search keyword or a specific phrase indicated by the target action specific information, and performs topic classification for each specific category using LDA from these plurality of words, and clusters the user U.

[0127] Further, the determination unit 32 can also extract a plurality of words from the content viewed by the user U who has performed a search using a search query including a specific search keyword or a specific phrase indicated by the target action specific information, perform topic classification for each specific category from these plurality of words using LDA, and cluster the user U.

[0128] The clustering by the determination unit 32 is not limited to the above-described example. For example, the determination unit 32 can cluster the user U using k-means clustering, hierarchical clustering, DBSCAN (Density-Based Spatial Clustering of Applications with Noise), etc., or can cluster the user U using a learned model such as a naive Bayes classifier, a support vector machine, a neural network, a generative AI, or can cluster the user U based on rules such as using a dictionary for each topic.

[0129] For example, the determination unit 32 inputs information of a specific category for each specific category into a learned model, and causes the learned model to execute topic classification of the specific category on the learned model. Then, the determination unit 32 clusters the user U by classifying the user U who has used a specific category into the topic corresponding to that specific category.

[0130] The determination unit 32 classifies the user U who has used a specific category into each of a plurality of different topics. For example, the user U can also be classified into the topic with the largest number of uses of a specific category.

[0131] The determination unit 32 extracts, as target topics, topics in which the ratio of the user U is equal to or higher than a threshold based on the ratio of the user U classified into each topic, and determines a persona corresponding to such target topics. The threshold is 5%, but is not limited to such an example.

[0132] The determination unit 32 determines a persona corresponding to the target topic based on, for example, the attribute information of the user U classified into the target topic and the target topic information which is the information of the target topic.

[0133] For example, the determination unit 32 determines, as the information of the persona corresponding to the target topic, information including the attribute with the highest ratio among the attributes of the user U classified into the target topic and the information of a specific category classified into the target topic. The attributes included in the information of the persona are indicated by, for example, gender and age, but are not limited to such examples. For example, in addition to or instead of gender and age, it may be indicated by family composition, annual income, place of residence, occupation, etc.

[0134] Further, the determination unit 32, for example, rounds the units digit of the ratio of the users U belonging to the topic with the ratio of the user U being equal to or greater than the threshold value, and then divides the value by 10 to obtain the number of personas corresponding to the topic, but is not limited to such an example.

[0135] When the number of personas corresponding to the same target topic is one, the determination unit 32 sets the attribute with the highest ratio among the attributes of the user U classified into the target topic as the attribute of the persona. Further, when the number of personas corresponding to the same target topic is two or more, the determination unit 32 sets each of the two or more attributes in the order of the attributes with the highest ratio among the attributes of the user U classified into the target topic as the attribute of the corresponding persona among the two or more personas.

[0136] The target topic information included in the information of the persona is information indicating the target topic, but may be information obtained by summarizing the information of a specific category classified into the target topic using a text generation AI or the like.

[0137] The text generation AI is, for example, a language model trained to estimate and output the next token from the input token sequence, and is, for example, a transformer-based model or an RNN-based model.

[0138] Transfer-based models include, for example, GPT, BARD, etc., but are not limited to such examples. RNN-based models include, for example, RWKV, etc., but are not limited to such examples. Note that it is desirable to perform learning so that the input information is not used as a new answer, thereby concealing information such as the input personal information.

[0139] For example, assume that the topics to be classified for user U are topics 1 to 10, and the target topics are topic 2, topic 4, topic 5, and topics 7 to 10. Also assume that the number of personas corresponding to topic 4, topic 9, and topic 10 is 2.

[0140] In this case, the determination unit 32 determines, for example, 10 personas such as a 30-year-old woman with topic 2 as the background, a 20-year-old woman with topic 4 as the background, a 30-year-old woman with topic 4 as the background, a 20-year-old woman with topic 5 as the background, a 30-year-old woman with topic 7 as the background, a 30-year-old woman with topic 8 as the background, a 30-year-old woman with topic 9 as the background, a 40-year-old woman with topic 9 as the background, a 20-year-old woman with topic 10 as the background, and a 30-year-old woman with topic 10 as the background.

[0141] [3.3.4. Generation unit 33] The generation unit 33 generates various information. For example, the generation unit 33 causes the generation AI to generate information indicating a discussion on a specific issue by a plurality of personas based on the information indicating the plurality of personas determined by the determination unit 32. The generation AI is the above-described text generation AI or the above-described multimodal AI.

[0142] For example, the generation unit 33 inputs, as input information to the generation AI, information including instruction information instructing a discussion on a specific issue by a plurality of personas determined by the determination unit 32, and causes the generation AI to generate information indicating a discussion by the plurality of personas.

[0143] A specific topic is the topic indicated by the topic information included in the discussion information request, but is not limited to such examples. For example, the generation unit 33 can also input information including information instructing the generation of a topic into the generation AI as input information, and cause the generation AI to generate topic information.

[0144] The instruction information includes, for example, topic information indicating a specific topic, persona information including information of each of a plurality of personas, the past speech history of the persona, output instruction information instructing the selection of the persona to speak from among the plurality of personas indicated by the persona information and the output of speech to the previously selected persona, and the like.

[0145] The generation unit 33 repeats the process of causing the generation AI to output output information including information indicating the persona selected as the next speaker and information indicating the speech of the persona selected as the current speaker by inputting the information including the instruction information into the generation AI as input information. Thereby, the generation unit 33 can cause the generation AI to generate discussion information, which is information indicating a discussion by a plurality of personas. Note that when the repetition count reaches a predetermined number of times, the generation unit 33 ends the generation of discussion information by the generation AI.

[0146] When the generation AI is GPT-4 of OpenAI, the topic information indicating a specific topic, the persona information including information of each of a plurality of personas, and the output instruction information, etc. are set as information of the system message, and the information indicating the selected persona and the past speech history of the persona are set as the user prompt, but is not limited to such examples.

[0147] In addition, the instruction information includes, for example, information such as the string "Discussions will be held among the following 10 users. The details of the topic and the 10 users are as follows.", enabling the generation AI to appropriately generate information indicating discussions by multiple personas. The detailed information of the users is persona information, which includes topic information, attribute information, etc. as described above. Note that the persona information may include, for example, information on categories with high usage frequency by user U instead of topic information.

[0148] Furthermore, the generation unit 33 can also input, as input information to the generation AI, information that further includes the end condition information included in the discussion information request in the instruction information. The end condition information is information indicating the end condition of the discussion regarding a specific topic.

[0149] In this case, the generation unit 33 includes, for example, information such as the string "Discussions will be held among the following 10 users. The topic, the end condition of the discussion, and the detailed information of the 10 users are as follows." in the instruction information, along with the topic information, the end condition information, the information indicating multiple personas, and the output instruction information, etc.

[0150] Moreover, the generation unit 33 can also input, as input information to the generation AI, information that further includes the constraint condition information included in the discussion information request in the instruction information. The constraint condition information is information indicating the constraint conditions for discussions by multiple personas and is information indicating the requirements of user M for the content of the discussion.

[0151] In this case, the generation unit 33 includes, for example, information such as the string "Discussions will be held among the following 10 users. The topic, the end condition of the discussion, and the detailed information of the 10 users are as follows." in the instruction information, along with the topic information, the end condition information, the information indicating multiple personas, and the output instruction information, etc.

[0152] Furthermore, the generation unit 33 can also input, as input information to the generation AI, information that further includes information instructing the progress of the discussion by a facilitator who summarizes the opinions of multiple personas.

[0153] In this case, the generation unit 33 includes, for example, role information, which is information indicating the role of the facilitator, in the instruction information. The role information includes, for example, information such as the string "#Facilitator\nYou are the facilitator who summarizes the conversation. Please assign turns to 10 users to speak. After listening to all users' opinions and when the conversation is summarized, the facilitator should summarize the conversation again and ask the users to speak again to dig deeper into the specific content.", but is not limited to such examples.

[0154] When the generation unit 33 instructs the facilitator to proceed with the discussion, the generation unit 33 includes, in the instruction information, information for instructing the facilitator to speak first. Thereby, the generation unit 33 can appropriately cause the generation AI to generate information indicating the discussion by a plurality of personas.

[0155] Also, when the reception unit 30 receives the question information, the generation unit 33 inputs, as input information to the generation AI, information including the question information in the instruction information. For example, assume that the topic is "Housing where child-rearing generations want to live (custom-built houses, detached houses for sale, condominiums, etc.)".

[0156] In this case, the user M can, by looking at the discussion among the personas, cause, for example, the information of the string "Human: After listening to everyone's opinions, I recognize that the surrounding environment is more important than the floor plan and functions of the house. Is there any mistake in this recognition?" to be transmitted from the terminal device 3 to the information processing device 1 as question information.

[0157] In this case, the generation unit 33 generates instruction information including the information of the string "Human: After listening to everyone's opinions, I recognize that the surrounding environment is more important than the floor plan and functions of the house. Is there any mistake in this recognition?" in the past speech history of the personas. Thereby, the generation unit 33 can allow the user M to participate in the discussion and can more appropriately proceed with the discussion by a plurality of personas.

[0158] In addition, when the reception unit 30 receives additional persona information, the generation unit 33 inputs information including the persona information with the additional persona information added as input information to the generation AI. The additional persona information includes, for example, the attribute information of the additional persona and the information of the topic corresponding to the additional persona.

[0159] Thereby, the generation unit 33 can input, as input information, information indicating a discussion by the additional persona indicated by the additional persona information and the plurality of personas described above to the generation AI as instruction information.

[0160] Note that the generation AI described above is arranged in an external information processing device, and the generation unit 33 causes the generation AI to generate and acquire information indicating a discussion via an API provided by the external information processing device, but is not limited to such an example. For example, the generation AI may be arranged in the information processing device 1.

[0161] [3.3.5. Creation unit 34] The creation unit 34 creates a mind map of the discussion on a specific topic by a plurality of personas based on the information indicating the discussion on the topic.

[0162] The creation unit 34 inputs, for example, information of the character string "Input the conversation of a certain discussion. Please summarize the conversation into a mind map. The output should be text data in a format corresponding to the syntax of mermaid.js." and information including the discussion information as input information to the generation AI, and causes the generation AI to output data used for generating the mind map.

[0163] Then, the creation unit 34 creates a mind map based on the text data output from the generation AI. Note that when the generation AI is a multimodal AI or an image generation AI, the creation unit 34 can also cause the generation AI to directly create a mind map. The image generation AI is, for example, StackGAN, AttnGAN, T2I with Transformers, DALL-E, etc., but is not limited to such examples.

[0164] Provider Unit 35 The provider unit 35 provides various types of information to the user M. For example, the provider unit 35 provides the discussion information generated by the generation unit 33 to the user M by transmitting the discussion information generated by the generation unit 33 to the terminal device 3 of the user M. The discussion information generated by the generation unit 33 is an example of information indicating a discussion on a specific topic by a persona, and the user M is an example of the target person.

[0165] For example, the provider unit 35 can provide the discussion information to the user M every time there is a statement by a persona or a facilitator. Thereby, the user M can grasp the progress of the discussion in real time.

[0166] FIG. 5 is a diagram showing an example of the discussion information provided to the user M by the provider unit 35 in the processing unit 12 of the information processing apparatus 1 according to the embodiment. FIG. 6 is a diagram showing another example of the discussion information provided to the user M by the provider unit 35 in the processing unit 12 of the information processing apparatus 1 according to the embodiment.

[0167] The discussion information 50 shown in FIG. 5 is, for example, when the user M is an employee of a travel company, the topic is "Travel measures for the child-raising generation", the end condition is "To summarize one specific travel measure for the child-raising generation based on the opinions of all", and the constraint condition is "Many of the child-raising generation stay for 1 night when traveling, and we want to implement measures to extend this to 2 nights or more. We want the measures to be considered based on the anxieties and needs of the child-raising generation regarding travel." The discussion information 50 shows the discussion information in this case. In the discussion information 50, the facilitator is progressing the discussion as the discussion progress controller.

[0168] The discussion information 60 shown in FIG. 6 is the discussion information when the user M is a local government official, the topic is "Child-raising support measures", the end condition is "To summarize three conditions for housing where the child-raising generation would like to live", and the constraint condition is "As a developer, I want to understand the housing requirements of the child-raising generation."

[0169] In the discussion information 60, the facilitator is conducting the discussion as the discussion progress supervisor, and the user M is asking questions such as "When I listen to everyone's opinions, I recognize that the surrounding environment is more important than the floor plan and functions of the house. Is there any mistake in this recognition? Also, do you think there is not much emphasis on the type of residence (such as condominiums or detached houses)? Please let me know your opinions on these." or "As children grow, what kind of houses are ideal for a house that can flexibly adapt to the changing lifestyle? Please let me know your opinions." and the discussion is progressing.

[0170] Also, the providing unit 35 provides the information of the mind map created by the creating unit 34 to the user M by transmitting the information of the mind map created by the creating unit 34 to the terminal device 3 of the user M.

[0171] FIG. 7 is a diagram showing an example of the information of the mind map provided to the user M by the providing unit 35 in the processing unit 12 of the information processing apparatus 1 according to the embodiment. The mind map 70 shown in FIG. 7 is generated by the creating unit 34 based on the discussion indicated by the discussion information shown in FIG. 5.

[0172] 〔4. Processing Procedure〕 Next, the information processing procedure by the processing unit 12 of the information processing apparatus 1 according to the embodiment will be described. FIG. 8 is a flowchart showing an example of the information processing by the processing unit 12 of the information processing apparatus 1 according to the embodiment.

[0173] As shown in FIG. 8, the processing unit 12 of the information processing apparatus 1 determines whether a discussion information request has been received (step S10). When the processing unit 12 determines that a discussion information request has been received (step S10: Yes), it acquires user information, which is the information of each user U who has performed the target action specified by the acquisition target information included in the discussion information request (step S11).

[0174] Subsequently, the processing unit 12 determines a plurality of different personas by clustering based on the user information of each user U acquired in step S11 (step S12). Then, the processing unit 12 causes the generation AI to generate information indicating a discussion on a specific issue by the plurality of personas determined in step S12 (step S13), and starts providing the discussion information to the user M (step S14).

[0175] When the process of step S14 ends, or when it is determined that no discussion information request has been received (step S10: No), the processing unit 12 determines whether question information has been received (step S15). When the processing unit 12 determines that question information has been received (step S15: Yes), the processing unit 12 adds the question information to the instruction information as the speech history (step S16).

[0176] When the process of step S16 ends, or when it is determined that no question information has been received (step S15: No), the processing unit 12 determines whether additional persona information has been received (step S17). When the processing unit 12 determines that additional persona information has been received (step S17: Yes), the processing unit 12 adds the additional persona information to the instruction information as the speech history (step S18).

[0177] When the process of step S17 ends, or when it is determined that no additional persona information has been received (step S17: No), the processing unit 12 determines whether the operation end timing has been reached (step S19). The processing unit 12 determines that the operation end timing has been reached, for example, when the power of the information processing apparatus 1 is turned off.

[0178] When the processing unit 12 determines that the operation end timing has not been reached (step S19: No), the process proceeds to step S10. When the processing unit 12 determines that the operation end timing has been reached (step S19: Yes), the process shown in FIG. 8 ends.

[0179] [5. Modification Example] In the above example, when a discussion information request is received by the reception unit 30, user information of each user U is acquired by the acquisition unit 31, a plurality of personas are determined by the determination unit 32, and discussion information is generated by the generation unit 33. However, the present invention is not limited to such an example.

[0180] For example, after a discussion information request is received by the reception unit 30, when predetermined conditions are satisfied, user information of each user U may be acquired by the acquisition unit 31, a plurality of personas may be determined by the determination unit 32, and discussion information may be generated by the generation unit 33.

[0181] The predetermined conditions may be, for example, a condition that the number of users U who have performed the target action indicated in the discussion information request is equal to or greater than a predetermined number, a condition that the number of users U who have performed the target action indicated in the discussion information request per unit time is equal to or greater than a predetermined number, and the like.

[0182] In addition, the generation unit 33 can also exclude some personas from the plurality of personas participating in the discussion. For example, when the reception unit 30 receives exclusion information, which is information indicating a persona to be excluded by the user M of the terminal device 3, the generation unit 33 excludes the information of the persona indicated by the exclusion information from the instruction information, and can cause the generation AI to generate discussion information using the instruction information from which the information of the persona indicated by the exclusion information has been excluded.

[0183] In the above example, the generation unit 33 generates discussion information by repeatedly performing a process of causing the generation AI to output output information including information indicating a persona selected as the next speaker and information indicating the speech of the persona selected as the current speaker by inputting information including the instruction information as input information to the generation AI. However, the generation of discussion information is not limited to such an example.

[0184] For example, the generation unit 33 can also cause the generation AI to generate discussion information by inputting, as input information to the generation AI, instruction information including output instruction information for instructing discussion to a plurality of personas, topic information, end condition information, constraint condition information, persona information, and the like.

[0185] The output instruction information in this case is, for example, information of the character string "Conduct a discussion with the following 10 users. The details of the topic and the 10 users are as follows. Please output the content of the discussion conducted by the 10 users.", but is not limited to such an example. Also in this case, the generation unit 33 can include, in the instruction information, information for instructing the progress of the discussion by the facilitator, question information, additional persona information, and the like.

[0186] [[6. Hardware Configuration]] The information processing apparatus 1 according to the above-described embodiment is realized by a computer 80 having a configuration as shown in FIG. 9, for example. FIG. 9 is a hardware configuration diagram showing an example of a computer 80 that realizes the functions of the information processing apparatus 1 according to the embodiment. The computer 80 includes a CPU 81, a RAM 82, a ROM (Read Only Memory) 83, an HDD (Hard Disk Drive) 84, a communication interface (I / F) 85, an input / output interface (I / F) 86, and a media interface (I / F) 87.

[0187] The CPU 81 operates based on a program stored in the ROM 83 or the HDD 84 and controls each unit. The ROM 83 stores a boot program executed by the CPU 81 when the computer 80 is started up, a program depending on the hardware of the computer 80, and the like.

[0188] The HDD 84 stores a program executed by the CPU 81, data used by such a program, and the like. The communication interface 85 receives data from other devices via the network N (see FIG. 2) and sends it to the CPU 81, and transmits the data generated by the CPU 81 to other devices via the network N.

[0189] The CPU 81 controls output devices such as a display and a printer, and input devices such as a keyboard or a mouse via the input / output interface 86. The CPU 81 acquires data from the input device via the input / output interface 86. Further, the CPU 81 outputs the data generated via the input / output interface 86 to the output device.

[0190] The media interface 87 reads a program or data stored in the recording medium 88 and provides it to the CPU 81 via the RAM 82. The CPU 81 loads such a program from the recording medium 88 onto the RAM 82 via the media interface 87 and executes the loaded program. The recording medium 88 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc), a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0191] For example, when the computer 80 functions as the information processing apparatus 1 according to the embodiment, the CPU 81 of the computer 80 realizes the functions of the processing unit 12 by executing the program loaded onto the RAM 82. Further, the HDD 84 stores the data in the storage unit 11. The CPU 81 of the computer 80 reads and executes these programs from the recording medium 88, but as another example, these programs may be acquired from another device via the network N.

[0192] 〔7. Others〕 Also, among the respective processes described in the above embodiments, all or part of the processes described as being automatically performed can be manually performed, or all or part of the processes described as being manually performed can be automatically performed by known methods. Additionally, regarding the processing procedures, specific names, and information including various data and parameters shown in the above documents and drawings, they can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

[0193] Also, each component of each illustrated device is conceptually functional and does not necessarily have to be physically configured as shown in the figures. That is, the specific form of the distribution and integration of each device is not limited to that shown, and all or part of it can be functionally or physically distributed and integrated in any unit according to various loads, usage situations, etc.

[0194] For example, the above-described information processing apparatus 1 may be implemented by a terminal device and a server computer, or may be implemented by a plurality of server computers. Also, depending on the function, the configuration can be flexibly changed, such as by calling an external platform or the like through an API or network computing.

[0195] Also, the above-described embodiments and modification examples can be appropriately combined as long as the processing contents do not conflict.

[0196] 〔8. Effects〕 As described above, the information processing apparatus 1 according to the embodiment includes an acquisition unit 31, a determination unit 32, and a generation unit 33. The acquisition unit 31 acquires user information, which is information of each user U who has performed a specific action in an online service. The determination unit 32 determines a plurality of different personas by clustering based on the user information acquired by the acquisition unit 31. The generation unit 33 causes the generation AI to generate information indicating a discussion on a specific topic by the plurality of personas based on the information indicating the plurality of personas determined by the determination unit 32. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the topic.

[0197] Also, the user information includes information indicating the attributes of the user U and information related to specific actions performed by the user U. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the topic.

[0198] Also, the generation unit 33 inputs, as input information to the generation AI, information including instruction information for instructing a discussion on a specific topic by the plurality of personas determined by the determination unit 32, and causes the generation AI to generate information indicating a discussion by the plurality of personas. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the topic.

[0199] Also, the generation unit 33 inputs, as input information to the generation AI, information further including the past speech history regarding a specific topic by the persona. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the topic.

[0200] Also, the information processing apparatus 1 includes a reception unit 30 that receives topic information, which is information indicating a topic, and the generation unit 33 causes the generation AI to generate information indicating a discussion by the plurality of personas with the topic indicated by the topic information as a specific topic. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the topic.

[0201] In addition, the reception unit 30 receives end condition information indicating the end condition of the discussion on a specific issue, and the generation unit 33 inputs information that further includes the end condition information received by the reception unit 30 in the instruction information as input information to the generation AI. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the issue.

[0202] In addition, the reception unit 30 receives constraint condition information indicating the constraint conditions of the discussion content by a plurality of personas, and the generation unit 33 inputs information that further includes the constraint condition information received by the reception unit 30 in the instruction information as input information to the generation AI. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the issue.

[0203] In addition, the information processing apparatus 1 includes a provision unit 35 that provides information indicating the discussion on a specific issue by a persona to the target person, the reception unit 30 receives question information indicating a question by the target person, and the generation unit 33 inputs information that further includes the question information received by the reception unit 30 in the instruction information as input information to the generation AI. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the issue.

[0204] In addition, the reception unit 30 receives additional persona information indicating an additional persona that is a persona other than the plurality of personas, and the generation unit 33 inputs information that uses the additional persona indicated by the additional persona information received by the reception unit 30 and the plurality of personas to instruct the discussion as instruction information as input information to the generation AI. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the issue.

[0205] In addition, the generation unit 33 inputs information that further includes information instructing the progress of the discussion by a facilitator who is a progress manager for summarizing the discussion by a plurality of personas in the instruction information as input information to the generation AI. Thereby, the information processing apparatus 1 can obtain a more appropriate discussion on the issue.

[0206] In addition, the information processing apparatus 1 includes a creation unit 34 that creates a mind map 70 of the discussion on a specific topic by a plurality of personas. As a result, the information processing apparatus 1 can more appropriately obtain information that makes it easier to grasp the content of the discussion on the topic.

[0207] In addition, the acquisition unit 31 acquires, as user information, information on each user U who has used the service in a specific category in the online Q&A service. As a result, the information processing apparatus 1 can more appropriately obtain the discussion on the topic.

[0208] In addition, the acquisition unit 31 acquires, as user information, information on each user U who has performed a search with a specific search query in the online search service. As a result, the information processing apparatus 1 can more appropriately obtain the discussion on the topic.

[0209] As described above, the embodiments of the present application have been described in detail with reference to the drawings. However, this is an example, and the present invention can be implemented in other forms in which various modifications and improvements are made based on the knowledge of those skilled in the art, including the aspects described in the column of the disclosure of the invention.

[0210] In addition, the "section (section, module, unit)" described above can be read as "means" or "circuit". For example, the acquisition unit can be read as an acquisition means or an acquisition circuit.

Description of Reference Numerals

[0211] 1, 4 Information processing apparatus 2, 3 Terminal device 10 Communication unit 11 Storage unit 12 Processing unit 20 User information storage unit 30 Reception unit 31 Acquisition unit 32 Decision unit 33 Generation unit 34 Creation unit 35 Provision unit 100 Information processing system N Network

Claims

1. An acquisition unit that acquires user information, which is information of each user who has performed a specific action in an online service; A determination unit that determines a plurality of different personas by clustering based on the user information acquired by the acquisition unit; A generation unit that causes a generation AI to generate information indicating a discussion on a specific topic by the plurality of personas based on information indicating the plurality of personas determined by the determination unit. An information processing apparatus characterized by the above.

2. The user information includes information indicating the attributes of the user and information regarding the specific action performed by the user. The information processing apparatus according to claim 1, characterized by the above.

3. The generation unit inputs information including instruction information instructing a discussion on the specific topic by the plurality of personas determined by the determination unit as input information to the generation AI, and causes the generation AI to generate information indicating a discussion by the plurality of personas. The information processing apparatus according to claim 2, characterized by the above.

4. The generation unit further inputs, as the input information, information including a past speech history regarding the specific topic by the persona to the generation AI. The information processing apparatus according to claim 3, characterized by the above.

5. Comprises a reception unit that receives topic information, which is information indicating a topic, The generation unit causes the generation AI to generate information indicating a discussion by the plurality of personas on the topic indicated by the topic information received by the reception unit as the specific topic. The information processing apparatus according to claim 3, characterized by the above.

6. The reception unit receives end condition information indicating an end condition of a discussion on the specific topic, The generation unit further inputs, as the input information, information including the end condition information received by the reception unit in the instruction information to the generation AI. The information processing apparatus according to claim 5, characterized by the above.

7. The reception unit receives constraint condition information indicating constraint conditions for the discussion content by the plurality of personas, The generation unit further inputs, as the input information, information including the constraint condition information received by the reception unit in the instruction information to the generation AI. The information processing apparatus according to claim 5 or 6, characterized by the above.

8. Comprises a provision unit that provides information indicating a discussion on the specific topic by the persona to a target person, The reception unit Receive question information indicating a question from the subject, The generation unit, Input, as input information to the generation AI, information including the question information received by the reception unit and further including the instruction information The information processing apparatus according to claim 5 or 6, characterized in that.

9. The reception unit, Receive additional persona information indicating an additional persona that is a persona other than the plurality of personas, The generation unit, Input, as input information to the generation AI, information including, as the instruction information, information instructing a discussion by the additional persona indicated by the additional persona information received by the reception unit and the plurality of personas The information processing apparatus according to claim 5 or 6, characterized in that.

10. The generation unit, Input, as input information to the generation AI, information including, in the instruction information, further information instructing the progress of a discussion by a facilitator who is a progress officer for summarizing the discussion by the plurality of personas The information processing apparatus according to claim 5 or 6, characterized in that.

11. Comprising a creation unit that creates a mind map of the discussion on the specific topic based on information indicating the discussion on the specific topic by the plurality of personas The information processing apparatus according to any one of claims 1 to 6, characterized in that.

12. The acquisition unit, Acquire, as the user information, information of each user who used the service in a specific category in the online Q&A service The information processing apparatus according to any one of claims 1 to 6, characterized in that.

13. The acquisition unit, Acquire, as the user information, information of each user who performed a search with a specific search query in the online search service The information processing apparatus according to any one of claims 1 to 6, characterized in that.

14. An information processing method executed by a computer, comprising: An acquisition step of acquiring user information, which is information of each user who performed a specific action in an online service; A determination step of determining a plurality of different personas by clustering based on the user information acquired in the acquisition step; A generation step of causing a generation AI to generate information indicating a discussion on a specific topic by the plurality of personas based on information indicating the plurality of personas determined in the determination step The information processing method, characterized in that.

15. An acquisition procedure for acquiring user information, which is information of each user who has performed a specific action in an online service, A determination procedure for determining a plurality of different personas by clustering based on the user information acquired by the acquisition procedure, A generation procedure for causing an AI to generate information indicating a discussion on a specific issue by the plurality of personas based on information indicating the plurality of personas determined by the determination procedure, and causing a computer to execute An information processing program characterized by the above.

Images