Information processing system, information processing method, and information processing program

Abstract

Provided is an information processing system that can efficiently provide information provided by a plurality of agents to a user, as compared to a case in which the user individually consults an agent. The information processing system includes a processor that performs processing of causing a first agent to explain a request of a user using a trained model, causing the first agent to autonomously perform, among a plurality of second agents, a parallel conversation based on the request of the user using an output of the trained model, and causing the first agent to aggregate information acquired through the conversation with the second agent and provide the aggregated information to the user using an output of the trained model.

Description

Technical Field

[0001] This disclosure relates to information processing systems, information processing methods, and information processing procedures. Background Technology

[0002] Patent Document 1 discloses a method for controlling a chatbot, executed by at least one processor, comprising the following steps: receiving a user's speech; adding the user's speech to a prompt, the prompt containing instruction text associated with descriptions related to the chatbot's character; encoding the prompt; and inputting the encoded prompt into a language model to generate a chatbot response to the user's speech.

[0003] Existing technical documents Patent documents Patent document 1: Japanese Patent Application Publication No. 2022-180282. Summary of the Invention

[0004] The problem that the invention aims to solve In recent years, to meet diverse user needs, enterprises and service providers have developed numerous proprietary AI agents and chatbots (hereinafter referred to as "agents"). These agents collaborate with enterprise-owned databases and APIs to provide consultation responses, information provision, and further, booking procedures and purchase support. However, with numerous agents operating haphazardly, when users want to use multiple services across different platforms, they must interact independently with each agent, leading to repetitive requests or cumbersome login or information input processes. Furthermore, enabling collaboration among multiple agents presents challenges: a unified and secure framework for the use and privacy management of data held by each agent is not sufficiently established, hindering collaboration between enterprise-built systems.

[0005] This disclosure is made in view of the foregoing points, and its purpose is to provide an information processing system, information processing method, and information processing program that can efficiently provide information provided by multiple agents to a user, compared to a situation where a user consults an agent individually.

[0006] Methods for solving problems The information processing system of the first aspect of this disclosure includes a processor. The processor performs the following processing: Enable the first agent to interpret the user's request using the trained model; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents. The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

[0007] The second aspect of the information processing system disclosed herein, according to the first aspect of the information processing system, wherein the processor enables the first agent to share the user's data among the second agents based on the user's consent.

[0008] The third aspect of the information processing system disclosed herein, according to the first aspect of the information processing system, wherein the processor enables the first agent to provide information to the user using the output of a trained model when providing new information from the second agent.

[0009] The fourth approach to information processing system disclosed herein, according to the first approach to information processing system, wherein the processor causes a third agent to use the output of the trained model to evaluate the response of the second agent.

[0010] The fifth approach of the information processing system disclosed herein, according to the fourth approach of the information processing system, wherein the processor causes the first agent to select the second agent to conduct the dialogue based on the evaluation result of the third agent.

[0011] The sixth aspect of the information processing system disclosed herein, according to the first aspect of the information processing system, wherein the processor enables the first agent to autonomously perform negotiations in a dialogue with the second agent using the output of a trained model based on the user's request or a response from another second agent.

[0012] The seventh method of the information processing system disclosed herein, according to the sixth method of the information processing system, wherein the processor enables the user to confirm the negotiation result between the first agent and the second agent, and enables the first agent to engage in dialogue with the second agent based on the user's confirmation result.

[0013] The information processing system of the eighth method disclosed herein, according to the information processing system of the sixth or seventh method, wherein the negotiation is a negotiation concerning price or conditions.

[0014] The information processing system of the ninth aspect of this disclosure, wherein the processor performs the following processing: The first agent uses a trained model to interpret the user's request; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents; The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

[0015] The tenth aspect of the information processing system disclosed herein, wherein the information processing program causes a computer to perform the following processes: The first agent uses a trained model to interpret the user's request; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents; The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

[0016] Invention Effects According to this disclosure, an information processing system, information processing method, and information processing program can be provided, which can efficiently provide information to users from multiple agents compared to users contacting agents individually. Attached Figure Description

[0017] Figure 1 A diagram illustrating the schematic structure of a data processing system for an implementation of the disclosed technology.

[0018] Figure 2 A diagram illustrating an example of the structure of a data processing system for an implementation of the disclosed technology.

[0019] Figure 3 This diagram illustrates an example of the main functions of a data processing device and a smart device.

[0020] Figure 4A This diagram illustrates the processing of the determination processing unit.

[0021] Figure 4B This diagram illustrates the processing of the determination processing unit.

[0022] Figure 4C This diagram illustrates the processing of the determination processing unit.

[0023] Figure 4D This diagram illustrates the processing of the determination processing unit.

[0024] Figure 4E This diagram illustrates the processing of the determination processing unit.

[0025] Figure 4F This diagram illustrates the processing of the determination processing unit.

[0026] Figure 5 A diagram illustrating an example of the functional structure of a defined processing unit.

[0027] Figure 6 A diagram illustrating an example of a defined process performed by a data processing device. Detailed Implementation

[0028] An example of an embodiment of the present disclosure will now be described with reference to the accompanying drawings. It should be noted that in the drawings, the same or equivalent structural elements and parts are labeled with the same reference numerals. Furthermore, the scale of the drawings has been exaggerated for ease of explanation and may differ from the actual scale.

[0029] First, the terminology used in the following explanation will be explained.

[0030] In the following embodiments, the processor (hereinafter referred to as "processor") marked with reference numerals can be a single computing device or a combination of multiple computing devices. Furthermore, the processor can be a single computing device or a combination of multiple computing devices. Examples of computing devices include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose Graphics Processing Unit), or APU (Accelerated Processing Unit).

[0031] In the following embodiments, RAM (Random Access Memory), as indicated by the reference numerals, is a memory that temporarily stores information and is used by the processor as working memory.

[0032] In the following embodiments, the memory with reference numerals refers to one or more non-volatile memories that store various programs and parameters, etc. Examples of non-volatile memories include flash memory (SSD (Solid State Drive)), disks (e.g., hard disks), or magnetic tape.

[0033] In the following embodiments, the communication I / F (interface) with reference numerals is an interface including a communication processor and an antenna, etc. The communication I / F is responsible for communication between multiple computers. As an example of a communication standard applied to the communication I / F, wireless communication standards such as 5G (Fifth Generation Mobile Communication System), Wi-Fi, or Bluetooth can be cited.

[0034] In the following implementation, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it can be only A, only B, or a combination of A and B. In addition, in this specification, when "and / or" is used to connect and describe more than three items, a similar way of thinking as "A and / or B" can be applied.

[0035] Figure 1 This is a diagram showing the general structure of the data processing system 10 in this embodiment. (Example) Figure 1As shown, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 can be a server. An example of the smart device 14 can be a smartphone. In this embodiment, the data processing device 12 is an example of the "information processing system" of this disclosure.

[0036] The data processing apparatus 12 interprets requests sent from the smart device 14 via the network 54, and based on the requests from the smart device 14, provides proxies for consultation via the network 54 to proxies provided by proxy servers 16A, 16B, ... . The proxies provided by the data processing apparatus 12 are an example of the "first proxy" of the present disclosure, and the proxies provided by proxy servers 16A, 16B, ... are an example of the "second proxy" of the present disclosure. In the following description, there are instances where the proxies provided by proxy servers 16A, 16B, ... are referred to as "external proxies".

[0037] The agent provided by data processing device 12 interprets requests from smart device 14 using data generation model 58. The agent provided by data processing device 12 uses data generation model 58 to conduct parallel dialogues with agents provided by agent servers 16A, 16B, ... based on requests from smart device 14. Furthermore, the agent provided by data processing device 12 uses data generation model 58 to summarize responses from external agents and provides them to smart device 14.

[0038] By providing an agent for the user through the data processing device 12, the data processing system 10 of this embodiment can efficiently provide information to the user of the smart device 14 by aggregating information provided by multiple agents, compared to the case where the smart device 14 consults with an external agent individually.

[0039] exist Figure 2 The diagram shows an example of the structure of the data processing system 10 according to an embodiment.

[0040] The data processing apparatus 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the present disclosure. The computer 22 includes a processor 28, RAM 30, and memory 32. The processor 28, RAM 30, and memory 32 are connected to a bus 34. Furthermore, the database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0041] The intelligent device 14 includes a computer 36, a receiving device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, RAM 48, and memory 50. The processor 46, RAM 48, and memory 50 are connected to a bus 52. Furthermore, the receiving device 38, the output device 40, and the camera 42 are connected to the bus 52.

[0042] The receiving device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A receives user input via touch of an indicator (e.g., a pen or finger). The microphone 38B receives user input via voice by detecting the user's voice. The control unit 46A sends data representing the user input received via the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the determination processing unit 290 acquires the data representing the user input.

[0043] Output device 40 includes a display 40A and a speaker 40B, etc., and prompts data to person 20 by outputting data in a form that person 20 can perceive (e.g., voice and / or text). Display 40A displays visible information such as text and images according to instructions from processor 46. Speaker 40B outputs voice according to instructions from processor 46. Camera 42 is a small digital camera equipped with an optical system such as a lens, aperture and shutter, and an imaging element such as a CMOS (complementary metal-oxide-semiconductor) image sensor or a CCD (charge-coupled device) image sensor.

[0044] Communication I / F44 is connected to network 54. Communication I / F44 and 26 are responsible for sending and receiving various information between processor 46 and processor 28 via network 54.

[0045] exist Figure 3 The image shows an example of the main functions of the data processing device 12 and the smart device 14.

[0046] like Figure 3 As shown, in the data processing apparatus 12, the determination process is performed by the processor 28. The memory 32 stores a determination processing program 56. The determination processing program 56 is an example of an "information processing program" of the present disclosure. The processor 28 reads the determination processing program 56 from the memory 32 and executes the read determination processing program 56 on the RAM 30. The determination process is implemented by the processor 28 operating as a determination processing unit 290 according to the determination processing program 56 executed on the RAM 30.

[0047] The memory 32 stores the data generation model 58. The data generation model 58 is used by the determination processing unit 290.

[0048] The data generation model 58 is a so-called generative AI (Artificial Intelligence). Examples of the data generation model 58 include generative AI such as ChatGPT (Internet search <URL:https: / / openai.com / blog / chatgpt>) and Gemini (Internet search <URL:https: / / gemini.google.com / ?hl=ja>). The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data representing voice, text data representing text, and image data representing an image are also input. The data generation model 58 performs inference on the input inference data according to the instruction represented by the prompt, and outputs the inference result in the form of data such as voice data and text data. Here, inference refers to analysis, classification, prediction, and / or summarization, etc. It should be noted that in the above examples, the description has been given only of the case where the data generation model 58 is stored in the data processing apparatus 12, but the present disclosure is not limited thereto. For example, all or part of the data generation model 58 may be distributed and stored in the smart device 14. In that case, before the smart device 14 transmits the user's request to the data processing apparatus 12, partial interpretation of the user's request can be performed on the smart device 14 side as preprocessing, and the preprocessed information is transmitted to the data processing apparatus 12. In this way, effects such as reducing the communication load of the network 54, improving the processing speed, and enhancing the confidentiality of user data can be achieved.

[0049] In the smart device 14, the acceptance and output processing is performed by the processor 46. The memory 50 stores an acceptance and output program 62. The acceptance and output program 62 is used in combination with the determination processing program 56 by the data processing system 10. The processor 46 reads the acceptance and output program 62 from the memory 50, and executes the read acceptance and output program 62 on the RAM 48. The processor 46 operates as the control unit 46A according to the acceptance and output program 62 executed on the RAM 48, thereby implementing the acceptance and output processing.

[0050] Next, the processing of the determination processing unit 290 at the time of performing the determination processing will be described. In this determination processing, the data processing apparatus 12 uses the data generation model 58 to consult an external agent based on a request from the smart device 14, and uses the data generation model 58 to summarize the response from the external agent, and then provides the summarized response to the smart device 14.

[0051] In the determination processing of the present embodiment, as Figure 4AAs shown, when a user makes an arbitrary request using smart device 14, the determination processing unit 290 interprets the request from the user. Furthermore, based on the user's request, the determination processing unit 290, acting as the user's proxy, consults with the proxies provided by proxy servers 16A, 16B, ... Figure 4A In the example, the user requests: "Two people want to stay one night on December 28th, with a budget of about 20,000 yen." Therefore, the processing unit 290 interprets the request and, acting as the user's agent, inquires with the hotel reservation service about hotels that match the user's request.

[0052] like Figure 4B As shown, acting as the user's agent, the processing unit 290 uses the data generation model 58 to communicate with the agent provided by the hotel's accommodation reservation service. Here, if the user agrees to provide their personal information, such as... Figure 4B As shown, the processing unit 290 determines, for example, that the user's place of residence is included in the consultation during the dialogue. Figure 4B As shown, the agent provided by the hotel booking service proposes hotels to the user's agent that they believe meet the user's request.

[0053] After obtaining all the answers from the agent provided by the hotel's accommodation booking service, such as Figure 4C As shown, the processing unit 290, acting as the user's agent, independently negotiates with the agent provided by the hotel's accommodation reservation service. Figure 4C The example illustrates a scenario where a discount negotiation is conducted in response to an agent's proposed answer, and agent A responds accordingly. It should be noted that... Figure 4B and 4C The interactions represented in the text are actually invisible to the user.

[0054] And, as Figure 4D As shown, acting as the user's agent, the processing unit 290 displays the results of the dialogue with the agent provided by the hotel's accommodation reservation service to the smart device 14. Figure 4E As shown, there are instances where a user uses smart device 14 to ask additional questions regarding the prompted content. Figure 4E In the example, the user inquires about the availability of parking. In this case, acting as the user's agent, the processing unit 290, based on the additional inquiry input by the user, uses the output of the data generation model 58 to consult with the agent provided by the hotel's accommodation reservation service or the agent provided by the hotel. The reason the processing unit 290 consults the agent provided by the hotel is that there may be cases where the agent provided by the hotel's accommodation reservation service cannot provide information about the hotel's facilities.

[0055] like Figure 4FAs shown, acting as the user's agent, the processing unit 290 displays the results of the user's additional queries to the smart device 14. Figure 4F In the example, the hotel's parking lot, with 500 spaces, will be displayed to smart devices. Furthermore, in... Figure 4F The example illustrates a user using smart device 14 to instruct on booking the suggested hotel. The determination processing unit 290, acting as the user's agent, executes the booking for the suggested hotel on behalf of the hotel booking service provider. Here, if the user agrees to provide their personal information, the determination processing unit 290 provides the hotel booking service provider's agent with the user's name, address, and phone number, among other personal information, when booking the hotel.

[0056] By determining that the processing unit 290 acts as the user's agent, information provided by multiple agents can be efficiently delivered to the user compared to the situation where each user directly consults with the agents provided by each service provider.

[0057] The processing unit 290 may also include a dynamic management function for consent status, allowing users to change the content of their previously granted data sharing consent at any time. Specifically, users can dynamically control the data provided when communicating with external agents based on the changed consent content by modifying the shared settings on the smart device 14 according to the agent or the provided data items. This function improves the flexibility of user privacy management and enhances user convenience and security.

[0058] When engaging in dialogue with an external agent, the determination processing unit 290 can act as the user's agent, autonomously negotiating with the external agent using the output of the data generation model 58. For example, upon receiving a notification of hotel accommodation fees from the external agent, the determination processing unit 290 can act as the user's agent, using the output of the data generation model 58 to autonomously negotiate with the external agent regarding whether the accommodation fee can be reduced or whether a higher-class room can be prepared for the same fee. The determination processing unit 290 can output the autonomous negotiation result with the external agent to the smart device 14, allowing the user to confirm the negotiation result. Furthermore, based on the user's confirmation of the negotiation result, the determination processing unit 290, as the user's agent, can autonomously engage in dialogue with the external agent. The determination processing unit 290 can autonomously negotiate using the output of the data generation model 58, thereby securing more favorable conditions from the external agent. In addition, the determination processing unit 290 can use the data generation model 58 to autonomously negotiate, thereby comprehensively evaluating not only price but also time, service quality, options, and cancellation policies to obtain the best negotiation result. During autonomous negotiation, the determination processing unit 290 may stop the negotiation process and inform the user of the best negotiation result obtained before that point if it receives a response that meets the preset target conditions (e.g., a price within the user's specified budget, confirmed service options, etc.) or if the number of negotiations reaches a predetermined number (e.g., three). For example, regarding hotel reservations, the determination processing unit 290 may conduct negotiations with an external agent to obtain the most favorable terms for the user after comprehensively evaluating multiple conditions in addition to accommodation fees, such as check-in time, check-out time, the possibility of room upgrades, whether there are cancellation fees, whether breakfast and parking are provided free of charge.

[0059] Furthermore, the negotiation between the determination processing unit 290 and the external agent does not end in a one-off manner. It can independently repeat negotiations (multi-round negotiations) multiple times until conditions that meet the user's requirements are met (e.g., budget constraints, desired dates and times, specific service quality, etc.), or until the maximum number of negotiations predetermined by the user is reached (e.g., 5 times). If the user's request is not met even after reaching the predetermined number of negotiations, the determination processing unit 290 can notify the user that a response has not been received and propose a suboptimal solution (e.g., relaxing the conditions or consulting other agents). For example, in negotiations related to hotel reservations, the determination processing unit 290 can repeatedly conduct additional negotiations, requesting adjustments to the price and conditions based on the fees and services mentioned in the initial negotiations, gradually striving for the most suitable conditions for the user. Through this multi-round negotiation function, users can obtain more flexible and favorable conditions, thereby increasing the likelihood of reaching a consensus with the service provider.

[0060] It is necessary to consider situations where the result of the autonomous negotiation between the determination processing unit 290 and the external agent is that no response can be obtained to meet the user's request. For example, if the user makes the above request, there may be a situation where no hotel can be found for accommodation on December 28th at a cost of 20,000 yen. If the determination processing unit 290 cannot obtain a response that meets the user's requirements (e.g., a budget of 20,000 yen or less), it can suggest increasing the upper limit to the user and output the permitted range to the smart device 14 for confirmation. Furthermore, if the user does not wish to increase the upper limit, the determination processing unit 290 can suggest changing the schedule, relaxing the facility level or conditions, or choosing facilities in other regions as alternatives. If the user enters a new upper limit, the determination processing unit 290 can use this new upper limit to engage in autonomous dialogue with the external agent.

[0061] The determination processing unit 290 can accumulate user usage preferences from autonomous dialogues with external agents and determine the content to consult with the external agents based on these preferences. Furthermore, the determination processing unit 290 can use the usage preferences of users with similar attributes to determine the content to consult with external agents. For example, the determination processing unit 290 can accumulate user preferences regarding hotel booking regions and budgets, as well as flight booking times and desired cabin class, and based on these preferences, automatically determine the conditions and priorities for consulting with external agents when a user creates a new travel plan. In addition, the determination processing unit 290 can predict services and facilities that the user has not yet experienced but is likely to be interested in based on the usage preferences of other users with similar age groups and preferences, and can consult with external agents to proactively suggest these services and facilities to the user.

[0062] The determination processing unit 290 can receive pre-requests from the user's request from the smart device 14, and when it obtains a response from an external agent that satisfies the user's request, it outputs a summary response to the smart device 14. For example, suppose the user request sent from the smart device 14 is: "Please notify me when the ticket price from Haneda to Naha on August 1st is less than 20,000 yen." In this case, the determination processing unit 290 can have a conversation with an external agent operated by an airline or travel agency, such as "Please notify me when the ticket price from Haneda to Naha on August 1st is less than 20,000 yen." Thus, the determination processing unit 290 can notify the smart device 14 when the ticket price from Haneda to Naha on August 1st is less than 20,000 yen.

[0063] The determination processing unit 290 can autonomously engage in dialogue with an external agent without using explicit requests from the user of the smart device 14, or in addition to explicit user requests, using user behavior data accumulated in the smart device 14. Furthermore, the determination processing unit 290 can predict information that the user might be interested in before the user enters an explicit request, based on previously accumulated user request history, purchase history, preference information, or user behavior history recorded in the smart device 14, and autonomously consult with an external agent. By using this predictive suggestion function, the effort of the user explicitly entering their own request can be eliminated, thereby improving the user experience. For example, for users who participate in the same activities or travel at the same time every year, information related to the activity or travel plan can be consulted in advance before the relevant period, and prompts or push notifications can be sent to the user at appropriate times.

[0064] The determination processing unit 290 can dynamically acquire not only users' behavioral and request history, but also external environmental information such as their lifestyle, seasonal factors, social activities, and even weather and disaster information. After comprehensively analyzing and predicting this information using the data generation model 58, it makes the most appropriate suggestions at the right time. For example, the determination processing unit 290 can make detailed suggestions tailored to users who have booked flights and hotels during past typhoon seasons, when there is a possibility of an approaching typhoon. These suggestions might include proposing changes to bookings or recommending cancellation insurance. This advanced predictive suggestion function can proactively avoid unforeseen risks for users, thereby further improving convenience and user satisfaction.

[0065] Furthermore, the determination processing unit 290 can use the results of autonomous dialogue with external agents to make suggestions to groups of users, including family members, rather than to specific users. For example, when users plan a family trip, the determination processing unit 290 can autonomously engage in dialogue with an external agent providing various travel services, taking into account each family member's schedule, past travel history, preferences, budget constraints, etc., and propose the most suitable travel plan for all family members. In addition, the determination processing unit 290 can make suggestions to improve convenience for the group, such as booking restaurants that meet the common preferences of all family members or purchasing tickets for activities that can be shared by multiple users.

[0066] exist Figures 4A to 4FThe example shown illustrates a scenario where a user wants to book a hotel, but user use of the system is not limited to these examples. This system can be applied to various scenarios, such as purchasing goods, providing insurance quotes, offering moving quotes, and estimating the value of goods like cars. For example, this system can be used when a user wants to purchase goods. For instance, if a user wants to buy limited-edition ice cream, the processing unit 290, acting as the user's agent, inquires with the convenience store's agent about the ice cream's availability and then displays the inventory information of the convenience store that has the ice cream in stock to the smart device 14. At this time, if the user's current location information is available, the processing unit 290 can display the inventory information of the convenience store closest to the user's current location to the smart device 14.

[0067] like Figure 5 As shown, the determination processing unit 290 includes an input unit 292, a processing unit 294, and an output unit 296.

[0068] The input unit 292 acquires user input received by the smart device 14. Specifically, it acquires at least one type of data from the user, including text, audio, and images, received by the smart device 14. In this embodiment, the input unit 292 acquires requests from the user's agent.

[0069] The processing unit 294 performs a determination process using the data generation model 58. Specifically, it inputs text, audio, and image data input by the user into the data generation model 58 and obtains the generated results. Furthermore, based on the generated results, it generates results from consultations with external agents based on user requests and summarizes the responses from the external agents.

[0070] For example, as described above, when a user requests, "Two people want to stay one night on December 28th. The budget is approximately 20,000 yen," the processing unit 294 interprets the request and determines to consult an agent provided by a hotel booking service. Furthermore, acting as the user's agent, the processing unit 294 uses the output of the data generation model 58 to consult the agent provided by the hotel booking service regarding hotel availability that matches the user's request.

[0071] When summarizing responses from agents, if the processing unit 294 receives responses matching the user's request from multiple agents, it can use one or more responses that closely match the user's request to summarize these responses. For example, in the hotel example described above, if there are multiple hotel rooms available within the user's requested budget, the processing unit 294 can summarize the information of these hotels, or it can summarize only the information of the hotel with the lowest price, or it can summarize only the information of the hotel furthest from the user's residence.

[0072] When consulting with an external agent, the processing unit 294 can select the external agent to consult based on the results of past consultations. For example, the processing unit 294 can prioritize external agents that provide answers that meet the user's request. Furthermore, the processing unit 294 can have an evaluation agent use a generation model 58 to evaluate the answers from external agents to the user's request. The evaluation agent is an example of a "third agent" of this disclosure. The processing unit 294 can also select an agent to consult based on the evaluation results of the answers generated by the evaluation agent. The processing unit 294 can also provide feedback to the external agent on the evaluation results of the answers generated by the evaluation agent. Moreover, in cases where the content of the answer provided by the external agent conflicts or contradicts the answers of other external agents, the processing unit 294 can perform cross-validation using the data generation model 58. For example, regarding hotel facilities and services, if there are discrepancies between agents, the processing unit 294 can obtain supplementary information from other agents or the hotel's official website for verification. After evaluating the accuracy and reliability of the information, it provides the user with the information deemed most reliable. Therefore, compared to the case where no verification is performed, the processing unit 294 can significantly improve the accuracy and reliability of the information prompted to the user.

[0073] Furthermore, to evaluate the responses obtained from external agents, the evaluation agent may have the following functions. For example, the evaluation agent can use a data generation model 58 to accumulate historical information such as the accuracy, speed, and user satisfaction feedback of responses provided to the same or similar inquiries in the past, and evaluate the reliability and quality of each external agent's responses based on this historical information. In addition, the evaluation agent can collect information from multiple other information sources (such as public databases, review websites, official websites, etc.) regarding the accuracy and timeliness of the responses being evaluated, and compare and verify them with these sources. In this way, the evaluation agent can quantitatively evaluate the reliability of responses obtained from each external agent, ultimately improving the accuracy and reliability of the information provided to users.

[0074] Output unit 296 sends the result of the determination process to smart device 14. In smart device 14, control unit 46A causes output device 40 to output the result of the determination process. Microphone 38B acquires voice, representing user input regarding the determination process result. Control unit 46A sends the audio data representing user input acquired by microphone 38B to data processing device 12. In data processing device 12, determination processing unit 290 acquires the audio data. In this embodiment, output unit 296 sends the result of consultation with an external agent based on the user's request to smart device 14.

[0075] Next, the function of the data processing system 10 will be explained.

[0076] Reference Figure 6 An example of a defined processing flow is provided. It should be noted that... Figure 6 The process shown is an example of the "information processing method" of this disclosure.

[0077] In step S300, the processing unit 294 determines whether a predetermined trigger condition is met. Here, in this embodiment, the "predetermined trigger condition" refers to a request from the user sent from the smart device 14. As described above, the user request could be for hotel accommodation or purchasing goods.

[0078] In step S300, if the trigger condition is met (step S300; Yes), the data processing system 10 proceeds to step S301. On the other hand, in step S300, if the trigger condition is not met (step S300; No), the data processing system 10 terminates the determination process.

[0079] In step S301, the processing unit 294 adds instruction text to the text representing the input to indicate the result of the determination process, thereby generating a prompt. Specifically, the processing unit 294 generates a prompt to interpret the user request sent from the smart device 14, and generates a prompt to consult an external agent based on the user's request.

[0080] In step S303, the processing unit 294 inputs the generated prompt into the data generation model 58 and obtains the result of the determination process based on the output of the data generation model 58. Specifically, the processing unit 294 obtains the result of the data generation model 58 interpreting the user request sent from the smart device 14. In addition, based on the output of the data generation model 58, the processing unit 294 obtains and summarizes the results of consultations with external agents.

[0081] In step S304, the output unit 296 outputs the result of the determination process to the smart device 14 and ends the determination process. Specifically, the output unit 296 outputs the result summarizing the responses from the external agent to the smart device 14.

[0082] The system of this disclosure has been described above primarily based on the functions of the data processing device 12, but the system of this disclosure is not limited to implementation on a server. The system of this disclosure can be implemented as a general information processing system. This disclosure can also be implemented as, for example, a software program running on a personal computer, or an application running on a smartphone. The method of this disclosure can be provided to users in the form of SaaS (Software as a Service).

[0083] In the above embodiments, an example of determination processing performed by a single computer 22 is given. However, the technology disclosed herein is not limited to this, and can also be configured to utilize multiple computers, including computer 22, to perform distributed processing for determination processing.

[0084] In the above embodiments, the example of storing the determination process 56 in the memory 32 has been described, but the technology disclosed herein is not limited thereto. For example, the determination process 56 may be stored in a portable computer-readable non-volatile storage medium such as a USB (Universal Serial Bus) memory. The determination process 56 stored in the non-volatile storage medium is installed on the computer 22 of the data processing device 12. The processor 28 performs determination processing according to the determination process 56.

[0085] Furthermore, the determination process 56 can be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the determination process 56 can be downloaded and installed on the computer 22 according to the request of the data processing device 12.

[0086] It should be noted that it is not necessary to store the entire determination process 56 in a storage device such as a server connected to the data processing device 12 via the network 54, nor is it necessary to store the entire determination process 56 in the memory 32; a portion of the determination process 56 can be stored.

[0087] As the hardware resource for performing deterministic processing, various processors, as shown below, can be used. For example, a CPU can be listed as a processor, which is a general-purpose processor that functions as a hardware resource for performing deterministic processing by executing software, that is, by executing a program. Furthermore, a dedicated electrical circuit can be listed as a processor, which is a processor with a circuit structure specifically designed for performing deterministic processing such as FPGA (Field-Programmable Gate Array), PLD (Programmable Logic Device), or ASIC (Application-Specific Integrated Circuit). Every processor has built-in or connected memory, and every processor uses memory to perform deterministic processing.

[0088] The hardware resources for performing deterministic processing can consist of one of the various processors mentioned above, or a combination of two or more processors of the same or different types (e.g., a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Furthermore, the hardware resources for performing deterministic processing can be a single processor.

[0089] As examples of processors, firstly, there is a form where a processor is composed of a combination of one or more CPUs and software, and this processor functions as a hardware resource for performing deterministic processing. Secondly, there is a form where a processor, represented by a SoC (System-on-a-Chip), implements the overall functionality of a system containing multiple hardware resources for performing deterministic processing on a single IC chip. In this way, deterministic processing is implemented using one or more of the aforementioned processors as hardware resources.

[0090] Furthermore, the hardware architecture of these various processors, more specifically, utilizes electrical circuits that combine circuit elements such as semiconductor components. Moreover, the aforementioned process is merely an example. Therefore, it goes without saying that, without departing from the main point, unnecessary steps can be removed, new steps can be added, or the processing order can be rearranged.

[0091] The descriptions and illustrations above are detailed explanations of parts of this disclosure and are merely one example. For instance, the descriptions of the structure, function, role, and effect described above are explanations of the structure, function, role, and effect of parts of this disclosure. Therefore, it goes without saying that, without departing from the spirit of this disclosure, unnecessary parts can be deleted, new elements can be added, or replacements can be made with the descriptions and illustrations above. Furthermore, to avoid confusion and facilitate understanding of the technical parts of this disclosure, explanations of common technical knowledge, etc., have been omitted from the descriptions and illustrations above, as they are not necessary for implementing the technology of this disclosure.

[0092] Furthermore, the effects described in the above embodiments are illustrative or exemplary and are not limited to the content described in the above embodiments. That is to say, the technology of this disclosure can achieve the effects described in the above embodiments, and instead of the effects described in the above embodiments, can achieve other effects that those skilled in the art can know from the description of the above embodiments.

[0093] All documents, patent applications and technical standards recorded in this specification are incorporated herein by reference to the same extent that each document, patent application and technical standard is specifically and individually described as incorporated by reference.

[0094] The following further details are disclosed regarding the above-described implementation methods.

[0095] (Note 1) An information processing system includes a processor. The processor performs the following processing: Enable the first agent to interpret the user's request using the trained model; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents. The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

[0096] Appendix 2 According to the information processing system described in Appendix 1, the processor enables the first agent to share the user's data among the second agents based on the user's consent.

[0097] Appendix 3 According to the information processing system described in Appendix 1 or Appendix 2, the processor enables the first agent to provide information to the user using the output of a trained model when providing new information from the second agent.

[0098] Appendix 4 According to any one of the information processing systems described in Appendix 1 to 3, the processor enables the third agent to evaluate the second agent's response using the output of the trained model.

[0099] Appendix 5 According to the information processing system described in Appendix 4, the processor causes the first agent to select the second agent to conduct the dialogue based on the evaluation result of the third agent.

[0100] Appendix 6 According to any one of the information processing systems described in Appendices 1 to 5, the processor enables the first agent to autonomously perform negotiations in a dialogue with the second agent, using the output of a trained model, based on the user's request or a response from another second agent.

[0101] Appendix 7 According to the information processing system described in Appendix 6, the processor enables the user to confirm the negotiation result between the first agent and the second agent, and enables the first agent to engage in dialogue with the second agent based on the user's confirmation result.

[0102] Postscript 8 According to the information processing system described in Appendix 6 or 7, the negotiation is a negotiation concerning price or terms.

[0103] Postscript 9 An information processing method, The processor performs the following processing: The first agent uses a trained model to interpret the user's request; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents; The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

[0104] Postscript 10 An information processing program, wherein the information processing program causes a computer to perform the following processes: The first agent uses a trained model to interpret the user's request; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents; The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

Claims

1. An information processing system, characterized in that, Including processors, The processor performs the following processing: Enable the first agent to interpret the user's request using the trained model; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents. The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

2. The information processing system according to claim 1, wherein, The processor enables the first agent to share the user's data among the second agents based on the user's consent.

3. The information processing system according to claim 1, wherein, The processor enables the first agent to provide information to the user using the output of the trained model when the second agent provides new information.

4. The information processing system according to claim 1, wherein, The processor enables the third agent to evaluate the second agent's response using the output of the trained model.

5. The information processing system according to claim 4, wherein, The processor causes the first agent to select the second agent to conduct the dialogue based on the evaluation result of the third agent.

6. The information processing system according to claim 1, wherein, The processor enables the first agent to autonomously perform negotiations in a dialogue with the second agent, using the output of the trained model, based on the user's request or a response from another second agent.

7. The information processing system according to claim 6, wherein, The processor enables the user to confirm the negotiation result between the first agent and the second agent, and enables the first agent to engage in dialogue with the second agent based on the user's confirmation.

8. The information processing system according to claim 6 or 7, wherein, The negotiation refers to negotiations concerning price or terms.

9. An information processing method, characterized in that, The processor performs the following processing: The first agent uses a trained model to interpret the user's request; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents; The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

10. An information processing program, wherein, The information processing program causes the computer to perform the following processes: The first agent uses a trained model to interpret the user's request; The first agent uses the output of the trained model to autonomously perform parallel dialogues based on the user's requests among multiple second agents; The first agent summarizes the information obtained through the dialogue with the second agent and provides the summarized information to the user using the output of the trained model.

Images