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

Abstract

To propose suitable question information for a user. [Solution] The information processing device according to the present application includes a first generation unit, a determination unit, and a provision unit. The first generation unit generates a context vector corresponding to a conversation history between a user and a chatbot based on the conversation history. The determination unit determines whether the context vector generated by the first generation unit satisfies a predetermined condition. The provision unit provides the user with question information based on the determination result determined by the determination unit, which is to be suggested to the user.

Description

[Technical Field]

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

[0002] Conventionally, a method of calling up a generating AI (Artificial Intelligence) on a predetermined chat tool has been known (see Non-Patent Document 1 below). [Prior art documents] [Patent documents]

[0003] [Non-Patent Document 1] "Part 4: Making generative AI accessible from Slack," [online], Ubie Tech Blog, Zenn, [Retrieved October 21, 2025], Internet<URL:https: / / zenn.dev / ubie_dev / articles / 6680c44737048d> Summary of the Invention [Problem to be solved by the invention]

[0004] However, the above-mentioned conventional technology merely discloses a method for calling up a generation AI on a specified chat tool, and it is not necessarily possible to suggest question information suitable for a user on such a specified chat tool.

[0005] The present application has been made in view of the above, and aims to provide an information processing device, an information processing method, and an information processing program that can propose question information suitable for a user. [Means for solving the problem]

[0006] The information processing device of the present application is characterized by comprising: a first generation unit that generates a context vector corresponding to a conversation history between a user and a chatbot based on the conversation history; a judgment unit that judges whether the context vector generated by the first generation unit satisfies a predetermined condition; and a provision unit that provides the user with question information based on the judgment result judged by the judgment unit, the question information being suggested to the user. [Effects of the Invention]

[0007] According to one aspect of the embodiment, it is possible to provide an effect that question information suitable for a user can be proposed. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of a providing process executed by an information processing device according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of a chat screen according to the embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of a chat screen according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of a conversation history storage unit according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of a question information storage unit according to the embodiment. [Figure 7] FIG. 7 is a flowchart showing an example of the flow of the providing process executed by the information processing device according to the embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of the flow of the providing process executed by the information processing device according to the embodiment. [Figure 9] FIG. 9 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the information processing device. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, an information processing device, an information processing method, and an 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 device, the information processing method, and the information processing program according to the present application are not limited to these embodiments. Furthermore, the embodiments can be appropriately combined as long as the processing content is not contradictory. Furthermore, the same components in the following embodiments are denoted by the same reference numerals, and duplicated explanations will be omitted.

[0010] 1. An example of a provision process executed by an information processing device The provision process executed by the information processing device 100 will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of the provision process executed by the information processing device 100 according to the embodiment.

[0011] First, let us explain the premise. In the example of Figure 1, it is assumed that user U1 is using a chat tool (an example of a predetermined chat tool) for exchanging messages with other users via chat. In the following, such a chat tool may be referred to as an application AP.

[0012] For example, assume that user U1 uses application AP to exchange messages with a chatbot on the chat of application AP. Here, exchanging messages with the chatbot refers to, for example, a conversation between user U1 and the chatbot using a generation AI that outputs answer information (output information) corresponding to question information (input information) received from user U1.

[0013] The application AP may be provided by the business operator that manages the information processing device 100, or may be provided by a business operator different from the business operator that manages the information processing device 100.

[0014] [1-1. Information Processing Systems] First, the configuration of an information processing system 1 according to an embodiment will be described with reference to Fig. 1. As shown in Fig. 1, the information processing system 1 includes a terminal device 10, a generation AI server 20, and an information processing device 100. The terminal device 10, the generation AI server 20, and the information processing device 100 are connected to each other via a network N such as the Internet, so as to be able to communicate with each other via wired or wireless communication. Note that the information processing system 1 shown in Fig. 1 may include multiple terminal devices 10, multiple generation AI servers 20, and multiple information processing devices 100.

[0015] The terminal device 10 is an information processing device used by a user U1 to access web pages and application content displayed in a browser. For example, the terminal device 10 is a mobile phone such as a smartphone, a tablet terminal, a desktop PC (Personal Computer), a notebook PC, a PDA (Personal Digital Assistant), etc. In the example of FIG. 1, it is assumed that the user U1 uses a smartphone. It is also assumed that the user U1 has installed an application AP on the terminal device 10.

[0016] In the following, the web page or application content displayed in the browser may be referred to simply as content. In the following, the terminal device 10 may be regarded as the same as a user (e.g., user U1). In other words, in the following, the user may be read as the terminal device 10.

[0017] The generation AI server 20 is an information processing device that receives question information from the information processing device 100 and provides answer information corresponding to the question information, and is realized by, for example, a server device or a cloud system. For example, the information processing executed by the generation AI server 20 is realized by a generation AI that generates text information. The generation AI here is, for example, a large language model (LLM) that is trained to estimate and output the next token from an input token sequence. For example, the LLM is a transformer-based model or an RNN (recurrent neural network)-based model.

[0018] For example, an LLM is a model trained to output answer information corresponding to received question information, and is a language model that performs natural language processing such as GPT (Generative Pre-trained Transformer) or Transformer. Note that the generative AI may be a multimodal LLM that can also receive image information, audio information, etc. as input information.

[0019] Examples of the transformer-based model include, but are not limited to, GPT and BERT (Bidirectional Encoder Representations from Transformers). Examples of the RNN-based model include, but are not limited to, RWKV (Receptance Weighted Key Value).

[0020] Furthermore, the LLM may be trained to prevent question information input to the LLM from being used as new answer information, thereby concealing personal information and other information contained in the question information. The generation AI may also be an LLM that has undergone dedicated training (for example, fine tuning) to generate answer information.

[0021] For example, the generation AI server 20 receives question information and a prompt from the information processing device 100. Next, the generation AI server 20 inputs the question information to the generation AI based on the prompt and generates answer information to be output from the generation AI. Then, the generation AI server 20 provides the generated answer information to the information processing device 100. Note that the generation AI server 20 may be realized by an API (Application Programming Interface).

[0022] The information processing device 100 is an information processing device capable of communicating with various devices via a network N, and is realized by, for example, a server device or a cloud system. For example, the information processing device 100 generates a context vector corresponding to a conversation history between a user and a chatbot based on the conversation history. Next, the information processing device 100 determines whether the generated context vector satisfies a predetermined condition. Then, the information processing device 100 provides the user with question information based on the determination result, which is to be suggested to the user.

[0023] [1-2. An example of provision processing] Next, a specific example of the provision process executed by the information processing device 100 will be described with reference to Fig. 1. In the example of Fig. 1, it is assumed that a user U1 exchanges messages with a chatbot on a chat of an application AP.

[0024] In the following, first, an example of providing answer information to question information received from a user will be described using steps S1 to S4 shown in Fig. 1. Next, an example of providing a plurality of pieces of question information to be suggested to the user will be described using steps S5 to S11 shown in Fig. 1.

[0025] First, an example of providing answer information to question information received from a user using steps S1 to S4 will be described. As shown in FIG. 1, the information processing device 100 receives question information from the terminal device 10 (step S1). For example, assume that the user U1 wants to know what points to keep in mind when granting a license. In this case, the information processing device 100 receives question information such as "Please tell me what points to keep in mind when reaching an agreement with the other party at the time of signing a license contract" from the terminal device 10.

[0026] Next, the information processing device 100 generates answer information to the received question information. Specifically, the information processing device 100 generates a prompt for generating the answer information. Then, the information processing device 100 transmits the prompt for generating the answer information and the question information to the generation AI server 20 (step S2). In this case, the generation AI server 20 generates the answer information in accordance with the instructions included in the prompt. For example, the generation AI server 20 generates answer information that explains, by item, points to note that are expected when granting a license.

[0027] Then, the information processing device 100 receives the answer information from the generation AI server 20 (step S3). Subsequently, the information processing device 100 provides the answer information to the terminal device 10 (step S4).

[0028] Next, an example in which a plurality of pieces of question information to be proposed to a user are provided to the user using steps S5 to S11 will be described. As shown in Fig. 1, the information processing device 100 generates a context vector (step S5). Here, the timing for generating the context vector is, for example, the timing after answer information is provided to the user.

[0029] For example, the information processing device 100 generates a context vector corresponding to the conversation history of the user U1 based on the conversation history. For example, the information processing device 100 generates a context vector that is a context indicated by the conversation history and indicates the context of the usage and / or purpose of the generation AI.

[0030] Next, the information processing device 100 determines whether the context vector satisfies a predetermined condition (step S6). For example, the information processing device 100 determines whether the first context vector satisfies a predetermined condition based on the similarity between the first context vector generated in step S5 and a second context vector stored in a predetermined storage unit of the information processing device 100, the second context vector being different from the first context vector. The similarity here may be, for example, cosine similarity.

[0031] To explain this using a more specific example, if the similarity between the first context vector and the second context vector is equal to or greater than a predetermined threshold, the information processing device 100 determines that the first context vector and the second context vector are similar. On the other hand, if the similarity between the first context vector and the second context vector is less than the predetermined threshold, the information processing device 100 determines that the first context vector and the second context vector are not similar.

[0032] Here, it is assumed that a plurality of pieces of question information, including previously provided question information and newly generated question information, are provided to user U1. For example, it is assumed that the question information provided to user U1 includes previously provided question information at a predetermined ratio. To explain this in more specific terms, it is assumed that the predetermined ratio is 50%. In this case, when 10 pieces of question information are provided to user U1, 5 pieces of question information are previously provided question information and 5 pieces of question information are newly generated question information.

[0033] In such a case, it is determined that the first context vector and the second context vector are similar. At this time, the information processing device 100 selects past question information (step S7). Specifically, the information processing device 100 selects question information that has been registered in association with the second context vector and that has been provided to the user in the past. For example, the information processing device 100 selects question information that corresponds to the second context vector from among the question information stored in a predetermined storage unit.

[0034] Also, it is assumed that it is determined that the first context vector and the second context vector are not similar. At this time, the information processing device 100 generates new question information. Specifically, the information processing device 100 generates a prompt for generating new question information. For example, the prompt includes an instruction for generating question information that takes into consideration the business content and in-house rules of the business operator that manages the information processing device 100.

[0035] The information processing device 100 then transmits a prompt for generating new question information to the generation AI server 20 (step S8). In this case, the generation AI server 20 generates new question information in accordance with the instructions included in the prompt. The information processing device 100 then receives the new question information from the generation AI server 20 (step S9).

[0036] Next, the information processing device 100 provides the question information to the terminal device 10 (step S10). Specifically, the information processing device 100 provides the terminal device 10 with a plurality of pieces of question information in a manner that allows the user U1 to select from the plurality of pieces of question information. For example, the information processing device 100 provides the terminal device 10 with a plurality of pieces of question information, including question information that has been provided in the past and newly generated question information, in a manner that allows the user U1 to select from the plurality of pieces of question information.

[0037] To explain this in more detail, the information processing device 100 provides question information in the form of buttons selectable by the user U1 as a chat message transmitted by the application AP. The information processing device 100 also provides a predetermined message along with the question information to prompt the user U1 to select the question information.

[0038] Selectable modes will be described using an example of the chat C1 shown in FIG. 1. In the example of FIG. 1, the chat C1 includes question information provided by the information processing device 100. The chat C1 includes a predetermined message, such as "Are you ready to choose the next fun story? Come on, let's choose the continuation of the conversation together!", which prompts the user U1 to select question information. The question information is displayed in the form of selectable buttons in the chat C1. The question information displayed in the chat C1 is "How can AI improve business efficiency?", "What are some specific examples of introducing an AI agent?", "What are the benefits of a chatbot?", "What kind of problems can AI solve?", and "What new service ideas do you have using AI?" For example, the question information "How can AI improve business efficiency?" and "What are some specific examples of introducing an AI agent?" are assumed to be question information selected as past question information. The question information "What are the benefits of a chatbot?", "What kind of problems can AI solve?", and "What new service ideas do you have using AI?" are assumed to be question information generated as new question information.

[0039] For example, if user U1 clicks on any of the following questions, "How can AI be used to make work more efficient?", "What are some specific examples of introducing AI agents?", "What are the benefits of chatbots?", "What problems can AI be used to solve?", or "What new service ideas do you have using AI?", one of the questions will be displayed in the chat. Also, if user U1 clicks on "+5 more" in chat C1, five other questions will be displayed.

[0040] Then, the information processing device 100 associates the context vector and the question information in a predetermined storage unit and registers them (step S11). For example, the information processing device 100 associates the context vector generated in step S5 with the question information provided in step S10 and registers them in a predetermined storage unit.

[0041] In this way, the information processing device 100 can suggest question information suitable for the user. Conventionally, with generation AI available on chat, it has sometimes been difficult to provide answer information suitable for the user when there are differences between users in the way they ask questions, the quality of the question content, the amount of explanation provided when asking the question, etc. Therefore, the information processing device 100 actively suggests question information that user U1 is likely to ask next. This allows the information processing device 100 to anticipate user U1's thoughts and encourage them to ask suitable questions, rather than simply providing assistance such as an automatic reply. Furthermore, the information processing device 100 can provide multiple question information to deepen the dialogue between user U1 and the chatbot.

[0042] Furthermore, the information processing device 100 can utilize the currently used question information the next time it is used by correlating the context vector with the question information and registering them in a predetermined storage unit. In this way, the information processing device 100 does not need to generate new question information every time it provides question information, and therefore the number of accesses to the API of the generation AI server 20 can be reduced.

[0043] [2. Example of a chat screen] Next, an example of proposing question information will be described with reference to Figs. 2 and 3. Figs. 2 and 3 are diagrams showing an example of a chat screen according to an embodiment. Fig. 2 shows a chat screen C2 on an application AP. For example, the chat screen C2 is a screen showing an exchange between a user U1 and a chatbot. In the example of Fig. 2, the chat screen C2 includes a button B21 such as "Conversation suggestions from Z bot" in addition to the exchange between the user U1 and the chatbot. When the user U1 presses the button B21, the question information is displayed.

[0044] Fig. 3 shows a chat screen C2 on which question information is displayed. In the example of Fig. 3, on the chat screen C2 on which question information is displayed, a chat for suggesting question information and sent by a chatbot is displayed at the bottom.

[0045] For example, the chat screen C2 shown in Fig. 3 includes an area RC21. The area RC21 includes a predetermined message such as "What should we talk about next? Please choose the continuation of the conversation and enjoy it!", and buttons B31 to B36 corresponding to each piece of question information displayed in a button format.

[0046] In the example of FIG. 3, button B31 is a button corresponding to question information such as "I would like to confirm the impact of patent rights." Button B32 is a button corresponding to question information such as "Who owns the intellectual property?" Button B33 is a button corresponding to question information such as "I would like to reconfirm the terms of use." Button B34 is a button corresponding to question information such as "Should the limitation of liability be revised?" Button B35 is a button corresponding to question information such as "What are the criteria for selecting the applicable law?" Button B36 is a button for displaying five other question information such as "+ 5 other items." In this way, user U1 can easily select question information provided by the information processing device 100 according to the situation.

[0047] 3. Configuration of Information Processing Device Next, an example of the functional configuration of the information processing device 100 will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of the configuration of the information processing device 100 according to an embodiment. As shown in Fig. 4, the information processing device 100 has a communication unit 110, a storage unit 120, and a control unit 130.

[0048] (Regarding the communication unit 110) The communication unit 110 is realized by, for example, a network interface card (NIC), etc. The communication unit 110 is connected to a network N by wire or wirelessly, and transmits and receives information to and from various other devices.

[0049] (Regarding the storage unit 120) The storage unit 120 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 storage unit 120 also has a conversation history storage unit 121 and a question information storage unit 122.

[0050] (Regarding the conversation history storage unit 121) The conversation history storage unit 121 stores various information related to the conversation history between the user and the chatbot. Fig. 5 shows an example of the conversation history storage unit 121 according to the embodiment. Fig. 5 is a diagram showing an example of the conversation history storage unit 121 according to the embodiment.

[0051] In the example shown in FIG. 5, the conversation history storage unit 121 has items such as "conversation history ID (Identifier)", "user ID", and "conversation history".

[0052] "Conversation history ID" is an identifier that identifies the conversation history. "User ID" is an identifier that identifies the user associated with the "conversation history ID." "Conversation history" is information about the conversation history associated with the "conversation history ID."

[0053] For example, in Figure 5, "H1" identified by the conversation history ID has a user ID of "U1" and a conversation history of "CH1." Note that in the example shown in Figure 5, the conversation history, etc., is expressed by an abstract code such as "CH1," but the conversation history, etc., may also be in the form of a file containing numerical values, character strings, or various information indicating the conversation history, etc.

[0054] (Regarding the question information storage unit 122) The question information storage unit 122 stores question information and a context vector in association with each other. Fig. 6 shows an example of the question information storage unit 122 according to the embodiment. Fig. 6 is a diagram showing an example of the question information storage unit 122 according to the embodiment.

[0055] In the example shown in FIG. 6, the question information storage unit 122 has items such as "question information ID," "context vector," and "question information."

[0056] "Question information ID" is an identifier for identifying question information. "Context vector" is a context vector associated with "Question information ID." "Question information" is question information associated with "Question information ID."

[0057] For example, in Fig. 6, "Q1" identified by the question information ID has a context vector of "CV1" and question information of "QI1." Note that in the example shown in Fig. 6, the context vector etc. is expressed by an abstract code such as "CV1," but the context vector etc. may also be in the form of a numerical value, a character string, or a file containing various information indicating the context vector etc.

[0058] (Regarding the control unit 130) The control unit 130 is a controller, and is realized by, for example, a central processing unit (CPU), a micro processing unit (MPU), a graphics processing unit (GPU), or the like executing various programs (examples of information processing programs) stored in a storage device inside the information processing device 100 using a RAM as a work area. The control unit 130 is also a controller, and is realized by, for example, an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

[0059] As shown in Fig. 4, control unit 130 has a reception unit 131, a first generation unit 132, a provision unit 133, a second generation unit 134, a determination unit 135, a selection unit 136, and a registration unit 137, and realizes or executes the functions and actions of information processing described below. Note that the internal configuration of control unit 130 is not limited to the configuration shown in Fig. 4, and may be other configurations as long as they perform the information processing described below. Furthermore, the connection relationship between each processing unit included in control unit 130 is not limited to the connection relationship shown in Fig. 4, and may be other connection relationships.

[0060] (Regarding the reception unit 131) The receiving unit 131 receives various types of information. Specifically, the receiving unit 131 receives question information from the terminal device 10. Then, the receiving unit 131 stores the question information in the storage unit 120.

[0061] For example, suppose that the user wants to know what points to keep in mind when granting a license. In this case, the receiving unit 131 receives question information from the terminal device 10, such as "Please tell me what points to keep in mind when reaching an agreement with the other party when granting a license."

[0062] (Regarding the first generation unit 132) The first generating unit 132 generates a context vector corresponding to the conversation history based on the conversation history between the user and the chatbot. Then, the first generating unit 132 stores the generated context vector in the storage unit 120.

[0063] For example, the first generating unit 132 generates a context vector that is a context indicated by the conversation history and indicates the context of the usage and / or purpose of use of the generation AI.

[0064] (About the provider 133) The providing unit 133 provides various types of information. Specifically, the providing unit 133 provides the answer information stored in the storage unit 120 to the terminal device 10.

[0065] Furthermore, the providing unit 133 provides the user with question information that is based on the determination result determined by the determining unit 135 and is to be suggested to the user. Specifically, the providing unit 133 provides the terminal device 10 with a plurality of pieces of question information in a manner that allows the user U1 to select from the plurality of pieces of question information. For example, the providing unit 133 provides the terminal device 10 with a plurality of pieces of question information that include question information that has been provided in the past and newly generated question information in a manner that allows the user to select from the plurality of pieces of question information.

[0066] To explain this with a more specific example, the providing unit 133 provides the question information in the form of buttons selectable by the user as a chat transmitted by the application AP. In addition, the providing unit 133 provides a predetermined message together with the question information to prompt the user U1 to select the question information.

[0067] (Regarding the second generation unit 134) The second generating unit 134 generates various types of information. Specifically, the second generating unit 134 generates answer information to the received question information. Then, the second generating unit 134 stores the generated answer information in the storage unit 120.

[0068] For example, the second generation unit 134 generates a prompt for generating answer information. Then, the second generation unit 134 transmits the prompt for generating answer information and question information to the generation AI server 20. In this case, the generation AI server 20 generates answer information in accordance with the instructions included in the prompt. For example, the generation AI server 20 generates answer information that explains points to note that are expected to be considered when granting a license, broken down into each item. Then, the second generation unit 134 receives the answer information from the generation AI server 20. In this way, the second generation unit 134 generates answer information.

[0069] Furthermore, if the first context vector and the second context vector are not similar, the second generation unit 134 generates new question information and stores the generated new question information in the storage unit 120.

[0070] For example, the second generation unit 134 generates a prompt for generating new question information. Then, the second generation unit 134 transmits the prompt for generating the new question information to the generation AI server 20. In this case, the generation AI server 20 generates the new question information in accordance with the instructions included in the prompt. Then, the second generation unit 134 receives the new question information from the generation AI server 20. In this way, the second generation unit 134 generates the new question information.

[0071] (Regarding the determination unit 135) The determination unit 135 determines whether or not the context vector generated by the first generation unit 132 satisfies a predetermined condition. For example, the determination unit 135 determines whether or not the first context vector satisfies a predetermined condition based on the similarity between the first context vector stored in the storage unit 120 and generated by the first generation unit 132, and the second context vector stored in the question information storage unit 122 (an example of a predetermined storage unit).

[0072] To explain this using a more specific example, when the similarity between the first context vector and the second context vector is equal to or greater than a predetermined threshold, the determination unit 135 determines that the first context vector and the second context vector are similar. On the other hand, when the similarity between the first context vector and the second context vector is less than the predetermined threshold, the determination unit 135 determines that the first context vector and the second context vector are not similar.

[0073] (Regarding the selection unit 136) When the first context vector and the second context vector are similar to each other, the selection unit 136 selects question information that is registered in association with the second context vector and that has been previously provided to the user. For example, the selection unit 136 selects question information that corresponds to the second context vector from among the question information stored in the question information storage unit 122.

[0074] (Regarding Registration Section 137) The registration unit 137 associates the context vector generated by the first generation unit 132 with the question information provided to the user and registers them in the question information storage unit 122. For example, the registration unit 137 associates the context vector generated in step S5 of Fig. 1 with the question information provided in step S10 of Fig. 1 and registers them in the question information storage unit 122.

[0075] [4. Processing Procedure (1)] Next, the procedure of the provision process executed by the information processing device 100 according to the embodiment will be described with reference to Fig. 7. Fig. 7 is a flowchart showing an example of the flow of the provision process executed by the information processing device 100 according to the embodiment.

[0076] 7, the receiving unit 131 receives question information from the terminal device 10 (step S101). Specifically, if the receiving unit 131 has not received question information from the terminal device 10 (step S101; No), the receiving unit 131 waits until question information is received.

[0077] On the other hand, when the receiving unit 131 receives question information from the terminal device 10 (Step S101; Yes), the second generating unit 134 generates answer information (Step S102). Subsequently, the providing unit 133 provides the answer information generated by the second generating unit 134 (Step S103).

[0078] [5. Processing Procedure (2)] Next, the procedure of the provision process executed by the information processing device 100 according to the embodiment will be described with reference to Fig. 8. Fig. 8 is a flowchart showing an example of the flow of the provision process executed by the information processing device 100 according to the embodiment.

[0079] 8, the first generating unit 132 determines whether it is a predetermined timing (step S201). For example, the predetermined timing is a timing after answer information is provided to the user.

[0080] Specifically, when the first generating unit 132 determines that it is not the predetermined timing (step S201; No), it waits until it determines that it is the predetermined timing. On the other hand, when the first generating unit 132 determines that it is the predetermined timing (step S201; Yes), it generates a context vector (step S202).

[0081] Next, the determination unit 135 determines whether the context vector generated by the first generation unit 132 satisfies a predetermined condition (step S203). Specifically, when the determination unit 135 determines that the context vector satisfies the predetermined condition (step S203; Yes), the selection unit 136 selects past question information (step S204). For example, the selection unit 136 selects question information from the question information stored in the question information storage unit 122 based on the context vector generated by the first generation unit 132.

[0082] On the other hand, if the determining unit 135 determines that the context vector does not satisfy the predetermined condition (step S203; No), the second generating unit 134 generates new question information (step S205).

[0083] Then, the providing unit 133 provides the question information to be proposed to the user to the terminal device 10 (step S206). Subsequently, the registering unit 137 associates the context vector generated by the first generating unit 132 with the question information provided to the user, and registers the associated context vector in the question information storage unit 122 (step S207).

[0084] [6. Modifications] The information processing device 100 described above may be implemented in various different forms other than the above embodiment, so other embodiments of the information processing device 100 will be described below.

[0085] [6-1.LLM] In the above embodiment, an example has been described in which the generation AI server 20 has an LLM, but this is not limiting. For example, the information processing device 100 may have an LLM. In this case, the LLM may be an LLM created independently by the business operator that manages the information processing device 100.

[0086] [6-2.Priority] Furthermore, the providing unit 133 may provide a plurality of pieces of question information arranged in an order according to a predetermined priority. The priority here is a priority set so as to give a higher priority to past question information. For example, the priority indicates an order of increasing similarity between the first context vector generated by the first generating unit 132 and the second context vector stored in the question information storage unit 122.

[0087] For example, the priority order is assumed to be "How can AI improve business efficiency?", "What are some specific examples of introducing an AI agent?", "What are the benefits of a chatbot?", "What kind of problems can AI solve?", and "What new service ideas do you have using AI?". In this case, the providing unit 133 may provide the terminal device 10 with question information arranged in the order of "How can AI improve business efficiency?", "What are some specific examples of introducing an AI agent?", "What are the benefits of a chatbot?", "What kind of problems can AI solve?", and "What new service ideas do you have using AI?". This allows the providing unit 133 to provide question information in a manner that is highly convenient for the user.

[0088] The priority is not limited to the above modification, and may be a priority determined by the LLM when generating new question information. For example, the second generation unit 134 generates a prompt for generating new question information, the prompt including an instruction to generate the new question information in an order based on the priority. The second generation unit 134 then transmits the prompt to the generation AI server 20. In this case, the generation AI server 20 generates new question information arranged in an order based on the priority according to the instruction included in the prompt. The second generation unit 134 may then receive the new question information arranged in an order based on the priority from the generation AI server 20.

[0089] [6-3. Number of characters] Furthermore, the providing unit 133 may provide question information having a character count within a predetermined threshold in a manner that allows the user to select the question information. For example, the providing unit 133 may provide question information having a character count within a predetermined threshold that indicates a character count that is easy for the user to check in a manner that allows the user to select the question information. This allows the providing unit 133 to provide question information in a manner that is highly convenient for the user. Furthermore, the user can easily select the provided question information depending on the situation.

[0090] [6-4. Question Information] Although the example has been described in which the question information provided by the providing unit 133 includes question information provided in the past and newly generated question information, the present invention is not limited to this. For example, the question information provided by the providing unit 133 may include only question information provided in the past. Alternatively, the question information provided by the providing unit 133 may include only newly generated question information.

[0091] Furthermore, in the above-described example, when the question information provided by the providing unit 133 includes question information provided in the past and newly generated question information, the question information provided by the providing unit 133 includes question information provided in the past at a predetermined ratio. However, the present invention is not limited to this. The predetermined ratio may be any ratio. For example, the number of question information provided in the past in the question information provided by the providing unit 133 may be smaller than the number of newly generated question information. Furthermore, the number of question information provided in the past in the question information provided by the providing unit 133 may be larger than the number of newly generated question information.

[0092] [6-5. Context Vector] The context vector may also have the same number of dimensions as the number of use cases in which the generative AI is used. A use case here refers to a situation in which a user uses the AI, such as summarizing information, organizing information, or deepening their thoughts.

[0093] The context vector may also be a context vector indicating multiple contexts. For example, the context vector may be a context vector indicating a time-series context such as the order of questions, or a context of the purpose and / or application of the generation AI. In this case, for example, the similarity may be calculated using a context vector indicating the context of the purpose and / or application of the generation AI. Furthermore, the priority of question information may be calculated using a context vector indicating a time-series context.

[0094] The context vector may also be a context vector generated for each user.

[0095] [6-6. Application Examples] Furthermore, in the above embodiment, information processing using an application AP has been described, but is not limited to this. The information processing executed by the information processing device 100 according to the above embodiment may be, for example, information processing using a generation AI available on a web page without using an application AP. Furthermore, the information processing executed by the information processing device 100 according to the above embodiment may be, for example, information processing executed by an AI agent.

[0096] The application AP may be used within the company of a business that manages the information processing device 100. The application AP may be used outside the company of a business that manages the information processing device 100.

[0097] [7. Program] The information processing device 100 according to the above-described embodiment is realized, for example, by a computer 1000 configured as shown in Fig. 9. Fig. 9 is a hardware configuration diagram showing an example of the computer 1000 that realizes the functions of the information processing device 100. The computer 1000 is connected to an output device 1010 and an input device 1020, and has a configuration in which a calculation device 1030, a cache 1040, a memory 1050, an output IF (Interface) 1060, an input IF 1070, and a network IF 1080 are connected via a bus 1090.

[0098] The arithmetic device 1030 operates based on programs stored in the cache 1040 or memory 1050, programs read from the input device 1020, etc., and executes various processes. The cache 1040 is a memory device, such as a RAM, that temporarily stores data used by the arithmetic device 1030 for various calculations. The memory 1050 is a storage device in which data used by the arithmetic device 1030 for various calculations and various databases are registered, and is realized by a ROM (Read Only Memory), HDD (Hard Disk Drive), flash memory, etc.

[0099] The output IF 1060 is an interface for transmitting information to be output to an output device 1010 that outputs various types of information, such as a monitor or a printer, and is realized by a connector conforming to a standard such as USB (Universal Serial Bus), DVI (Digital Visual Interface), or HDMI (High Definition Multimedia Interface), etc. The input IF 1070 is an interface for receiving information from an input device 1020, such as a mouse, keyboard, or scanner, and is realized by, for example, a USB.

[0100] The input device 1020 may be a device that reads information from an optical recording medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), or a PD (Phase Change Rewritable Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory, etc. The input device 1020 may also be an external storage medium such as a USB memory.

[0101] The network IF 1080 receives data from other devices via the network N and sends it to the arithmetic device 1030, and also transmits data generated by the arithmetic device 1030 to other devices via the network N.

[0102] The arithmetic unit 1030 controls the output device 1010 and the input device 1020 via the output IF 1060 and the input IF 1070. For example, the arithmetic unit 1030 loads a program from the input device 1020 or the memory 1050 onto the cache 1040 and executes the loaded program.

[0103] For example, when the computer 1000 functions as the information processing device 100 , the arithmetic unit 1030 of the computer 1000 executes a program loaded onto the cache 1040 to realize the functions of the control unit 130 .

[0104] [8. Other] Furthermore, among the processes described in the above embodiments and modifications, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using known methods. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown in the drawings.

[0105] Furthermore, the components of each device shown in the figure are conceptual functional components and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0106] Furthermore, the above-described embodiments and modifications can be combined as appropriate within the scope of not causing any contradiction in the processing content.

[0107] Furthermore, the above-mentioned "section, module, unit" can be read as "means" or "circuit," etc. For example, a providing section can be read as providing means or providing circuit.

[0108] [9. Effects] As described above, the information processing device 100 according to the embodiment includes the first generation unit 132, the determination unit 135, and the provision unit 133. The first generation unit 132 generates a context vector corresponding to a conversation history based on a conversation history between a user and a chatbot. The determination unit 135 determines whether or not the context vector generated by the first generation unit 132 satisfies a predetermined condition. The provision unit 133 provides the user with question information based on the determination result determined by the determination unit 135, which is to be suggested to the user.

[0109] As a result, the information processing apparatus 100 according to the embodiment can propose question information suitable for the user.

[0110] Furthermore, in the information processing device 100 according to the embodiment, the judgment unit 135 judges whether the first context vector satisfies a predetermined condition based on the similarity between the first context vector generated by the first generation unit 132 and the second context vector stored in a predetermined storage unit (corresponding to the question information storage unit 122 according to the embodiment).

[0111] As a result, the information processing apparatus 100 according to the embodiment can propose question information suitable for the user.

[0112] Furthermore, in the information processing device 100 according to the embodiment, the judgment unit 135 judges whether the first context vector satisfies a predetermined condition based on the similarity between the first context vector generated by the first generation unit 132 and the second context vector stored in a predetermined storage unit (corresponding to the question information storage unit 122 according to the embodiment).

[0113] This allows the information processing apparatus 100 according to the embodiment to select question information that is suitable for the user.

[0114] In addition, in the information processing device 100 according to the embodiment, the determination unit 135 determines that the first context vector and the second context vector are similar when the similarity between the first context vector and the second context vector is equal to or greater than a predetermined threshold, and further includes a selection unit 136 that selects question information that has been registered in association with the second context vector and that has been previously provided to the user when the first context vector and the second context vector are similar.

[0115] As a result, the information processing apparatus 100 according to the embodiment can select question information suitable for the user from among question information previously provided to the user.

[0116] In addition, in the information processing device 100 according to the embodiment, the determination unit 135 determines that the first context vector and the second context vector are not similar if the similarity between the first context vector and the second context vector is less than a predetermined threshold, and further includes a second generation unit 134 that generates new question information if the first context vector and the second context vector are not similar.

[0117] As a result, the information processing apparatus 100 according to the embodiment can generate question information that is suitable for the user.

[0118] In the information processing device 100 according to the embodiment, the second generation unit 134 generates new question information using a predetermined generation AI.

[0119] As a result, the information processing apparatus 100 according to the embodiment can generate question information that is suitable for the user.

[0120] In the information processing device 100 according to the embodiment, the providing unit 133 provides a plurality of pieces of question information to the user.

[0121] As a result, the information processing apparatus 100 according to the embodiment can provide a plurality of pieces of question information that are suitable for the user.

[0122] Furthermore, in the information processing device 100 according to the embodiment, the providing unit 133 provides the user with a plurality of pieces of question information, including question information that has been provided in the past and newly generated question information.

[0123] As a result, the information processing apparatus 100 according to the embodiment can provide a plurality of pieces of question information that are suitable for the user.

[0124] Furthermore, in the information processing device 100 according to the embodiment, the providing unit 133 provides a plurality of pieces of question information in a manner that allows the user to select one.

[0125] As a result, the information processing apparatus 100 according to the embodiment can provide a plurality of pieces of question information that are suitable for the user in a selectable manner.

[0126] Furthermore, in the information processing device 100 according to the embodiment, the providing unit 133 provides a plurality of pieces of question information arranged in an order according to a predetermined priority.

[0127] As a result, the information processing device 100 according to the embodiment can provide question information in a manner that is highly convenient for the user.

[0128] Furthermore, in the information processing device 100 according to the embodiment, the providing unit 133 provides question information having a character count within a predetermined threshold in a manner that allows the user to select the question.

[0129] As a result, the information processing device 100 according to the embodiment can provide question information in a manner that is highly convenient for the user.

[0130] Furthermore, in the information processing device 100 according to the embodiment, the providing unit 133 provides question information in the form of a button that can be selected by the user as a chat message sent in a predetermined chat tool.

[0131] As a result, the information processing device 100 according to the embodiment can provide question information in a manner that is highly convenient for the user.

[0132] Furthermore, in the information processing device 100 according to the embodiment, the providing unit 133 provides question information in the form of a button that can be selected by the user as a chat message sent in a predetermined chat tool.

[0133] As a result, the information processing device 100 according to the embodiment can provide question information in a manner that is highly convenient for the user.

[0134] Furthermore, in the information processing device 100 according to the embodiment, the providing unit 133 provides a predetermined message together with the question information to prompt the user to select the question information.

[0135] As a result, the information processing device 100 according to the embodiment can provide question information in a manner that is highly convenient for the user.

[0136] Moreover, the information processing device 100 according to the embodiment further includes a registration unit 137 that registers the context vector generated by the first generation unit 132 and question information provided to the user in association with each other.

[0137] As a result, the information processing device 100 according to the embodiment can utilize the question information used this time the next time it is used. In this way, the information processing device 100 does not need to generate new question information every time it provides question information, so the number of accesses to the API of the generation AI server 20 can be reduced.

[0138] Although some of the embodiments of the present application have been described in detail above with reference to the drawings, these are merely examples, and the present invention can be implemented in other forms that include the embodiments described in the Disclosure of the Invention section and that have been modified and improved in various ways based on the knowledge of those skilled in the art. [Explanation of symbols]

[0139] N Network 1. Information Processing Systems 10 Terminal Equipment 20 Generative AI Server 100 Information processing device 110 Communications Department 120 Storage section 121 Conversation history memory unit 122 Question information storage section 130 Control Unit 131 Reception 132 1st generation part 133 Provision Department 134 Second generation part 135 Judgment section 136 Selection Section 137 Registration Department

Claims

1. a first generation unit that generates a context vector corresponding to a conversation history between a user and a chatbot based on the conversation history; a determination unit that determines whether the context vector generated by the first generation unit satisfies a predetermined condition; a provision unit that provides, to the user, question information based on the determination result determined by the determination unit, to be proposed to the user; An information processing device comprising:

2. The determination unit Whether or not the first context vector satisfies a predetermined condition is determined based on the similarity between the first context vector generated by the first generation unit and the second context vector stored in a predetermined storage unit.

2. The information processing apparatus according to claim 1, wherein:

3. The determination unit If the similarity between the first context vector and the second context vector is equal to or greater than a predetermined threshold, the first context vector and the second context vector are determined to be similar; The system further includes a selection unit that selects, when the first context vector and the second context vector are similar, question information that has been registered in association with the second context vector and that has been previously provided to the user.

3. The information processing apparatus according to claim 2, wherein:

4. The determination unit If the similarity between the first context vector and the second context vector is less than a predetermined threshold, it is determined that the first context vector and the second context vector are not similar to each other; The system further includes a second generation unit that generates new question information when the first context vector and the second context vector are not similar to each other.

3. The information processing apparatus according to claim 2, wherein:

5. The second generation unit Generate the new question information using a predetermined generation AI.

5. The information processing apparatus according to claim 4,

6. The providing unit A plurality of pieces of question information are provided to the user.

2. The information processing apparatus according to claim 1, wherein:

7. The providing unit The plurality of pieces of question information including previously provided question information and newly generated question information are provided to the user.

7. The information processing apparatus according to claim 6,

8. The providing unit The plurality of pieces of question information are provided in a manner that allows the user to select the pieces of question information.

7. The information processing apparatus according to claim 6,

9. The providing unit The plurality of pieces of question information are arranged in an order according to a predetermined priority and provided.

9. The information processing apparatus according to claim 8,

10. The providing unit The question information, the number of characters of which is within a predetermined threshold, is provided in a manner selectable by the user.

9. The information processing apparatus according to claim 8,

11. The providing unit The question information is provided in the form of a button selectable by the user as a chat sent in a predetermined chat tool.

9. The information processing apparatus according to claim 8,

12. The providing unit Providing a predetermined message to prompt the user to select the question information together with the question information 12. The information processing apparatus according to claim 11,

13. The system further includes a registration unit that associates and registers the context vector generated by the first generation unit with the question information provided to the user.

2. The information processing apparatus according to claim 1, wherein:

14. 1. A computer-implemented information processing method, comprising: a first generation step of generating a context vector corresponding to a conversation history between a user and a chatbot based on the conversation history; a determination step of determining whether the context vector generated by the first generation step satisfies a predetermined condition; a providing step of providing to the user question information based on the determination result determined in the determining step, the question information being proposed to the user; An information processing method comprising:

15. a first generation step of generating a context vector corresponding to a conversation history between a user and a chatbot based on the conversation history; a determination step of determining whether the context vector generated by the first generation step satisfies a predetermined condition; a provision step of providing to the user question information based on the determination result determined by the determination step, the question information being proposed to the user; An information processing program characterized by causing a computer to execute the above.

Images