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

Abstract

An information processing system according to one aspect of the present invention comprises: a plurality of service units that input and output data to and from each other and interact with a user; a PDS unit that stores input / output logs of the service units and supplies stored data to the service units after processing the data as necessary; and an evaluation unit that is included in a service section and evaluates other service units and the PDS unit.

Description

Information Processing System, Information Processing Apparatus, Information Processing Method, and Program

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

[0002] Conventionally, various services that can be used using digital devices such as smartphones and personal computers have become widespread. To enhance the added value of these services, it is necessary for users to trust the services and entrust the utilization of their own data. Non-Patent Document 1 points out that data on intervention in users by AI (Artificial Intelligence) is necessary for AI governance.

[0003] The Nikkei Electronic Edition, "What are the Risks of Personal AI? Interview with Professor Koichi Hashida of the University of Tokyo" (August 29, 2023) <Internet URL: https: / / www.nikkei.com / article / DGXZQOUC248N20U3A720C2000000 / >, Search Date: June 26, 2024

[0004] There are cases where a user may want to use common data in multiple services. There are also cases where a user may want to switch to a service that is considered better among services of the same type. In these cases, a technology that enables a user to easily transfer data flexibly between services while appropriately protecting the data and referring to the evaluation of each service is desired.

[0005] Therefore, an object of the present invention is to provide an information processing system, an information processing apparatus, an information processing method, and a program that can facilitate service improvement and switching between services.

[0006] An information processing system according to one aspect of the present invention comprises a plurality of service units and a PDS unit. The plurality of service units may include an evaluation unit, an external data analysis system, and an external data supply system. The PDS unit stores data including logs of input and output from the plurality of service units (input and output of user interaction, input and output between service units, and input from the PDS unit), and processes the data as needed before providing it to the service units. There may or may not be users. If there are users, the users may be individuals or organizations such as companies. There may also be service units that are also PDS units. For example, a service unit that interacts with a user may be a PDS unit. The evaluation unit evaluates the units being evaluated (service units and PDS units) and can recommend or automate the use of some of the units being evaluated that have high evaluations. In the following, without loss of generality, each service unit will be assumed to provide exactly one service. When the evaluation unit evaluates a service unit, it means that it evaluates a specific service provided by that service unit.

[0007] According to this embodiment, data such as input / output logs from multiple service units are stored in the PDS unit. By using this data, it becomes easier to improve services in accordance with service evaluations and to switch between services.

[0008] According to the present invention, it becomes easier to improve services and switch between services.

[0009] This figure shows an example of the schematic configuration of the information processing system 1 according to the embodiment. This figure shows a specific example of the functions of the service. This block diagram shows an example of the schematic configuration of the hardware of the user terminal 10 according to the embodiment. This flowchart shows an example of processing by the information processing system 1 according to the embodiment. This figure shows an example of the evaluation information screen according to the embodiment.

[0010] A preferred embodiment of the present invention will be described with reference to the attached drawings. (Note that in each drawing, components denoted by the same reference numerals have the same or similar components.)

[0011] (1) Overview Figure 1 is a diagram showing an example of the schematic configuration of the information processing system 1 according to the embodiment. The information processing system 1 is composed of, for example, a user terminal 10 and various servers 20 (20a, 20b, 20c...). The user terminal 10 has a service unit 11 and a PDS (Personal Data Store) unit 12. The service unit 11 includes an AI (Artificial Intelligence) unit 13 and an evaluation unit 14.

[0012] Users can utilize various services provided by the service unit 11 through the information processing system 1. The service unit 11 may be connected to various partner servers 20 (20a, 20b, 20c, ...) via a communication network such as the Internet. Each service unit 11 may cooperate with the partner servers 20, etc., to provide services. These services may be provided in physical environments such as schools, hospitals, hotels, restaurants, and railways. In other words, each service unit 11 may provide services that users utilize in those physical environments. In Figure 1, as an example, the information processing system 1 is shown to be configured including a user terminal 10 and various servers 20 (20a, 20b, 20c, ...), but each service unit 11 may consist only of a user terminal 10 without including a server 20.

[0013] The information processing system 1 may have, for example, the following features (1-1) and / or (1-2).

[0014] (1-1) Storage of Service Unit Input / Output Logs in the PDS Unit The information processing system 1 generates logs related to the input and output of each service unit 11 when the service provided by each service unit 11 is used. The logs may be generated through interaction between the user and the service unit 11, through input and output between service units 11, or through the supply of data from the PDS unit 12 and an external data supply system 15 to the service unit 11. The generated logs are stored in the PDS unit 12. Here, if the input and output data of the service unit is large, a summary of the data or metadata may be included in the log. Furthermore, the PDS unit 12 may allow the user to comprehensively and centrally store and manage their own data without entrusting it to others.

[0015] In the information processing system 1, data including logs generated by the use of various services is stored in the PDS unit 12. This makes it easy for users to use logs obtained from one service in another service, switch from one service to another, or use data acquired on one terminal on another terminal. In Figure 1, the PDS unit 12 is shown as being located in the user terminal 10, but the PDS unit 12 may be located outside the user terminal 10. Also, there may be multiple PDS units 12 within the information processing system 1. The data supply system 15 is located outside the information processing system 1 and may also be multiple units.

[0016] (1-2) Evaluation of services through data analysis There may be multiple evaluation units 14 in the information processing system 1. The information processing system 1 may be configured to evaluate the units to be evaluated (various service units 11 (14a, 14b, 14c, ...), PDS unit 12, data supply system 15, and data analysis system 16) by having the evaluation unit 14 analyze data such as logs stored in the PDS unit 12, and to disclose the evaluation results to the user or to recommend the use of the highly-rated units to be evaluated.

[0017] The evaluation results may be a summary of the logs or a score based on some evaluation scale. The log summary of the service unit 11 may be the trend of financial assets in the case of investment-related services, or the trend of health status in the case of health-related services. The evaluation scale can be arbitrarily set according to the content of the services provided by the service unit 11. Specifically, for purchasing-related services, it may be purchase history; for financial services, it may be financial assets; for health-related services, it may be health status; and for academic or employment-related services, it may be an evaluation from the perspective of academic performance or work performance. Through these analyses, it is possible to evaluate, for example, how much economic, social, or health benefit each service unit 11 brings to the user, or is likely to bring. Furthermore, by disclosing the evaluation of the evaluated unit (service unit 11, PDS unit 12, data supply system 15, and data analysis system 16) to the user, the user can choose a better evaluated unit, and thus, according to market principles, the quality of services provided by the information processing system 1 can be improved throughout society, and user trust in the service can be fostered and maintained.

[0018] (2) The functional configuration of the information processing system 1 will be described with reference to the functional configuration diagram 1.

[0019] (2-1) User terminal 10 The user terminal 10 is an example of an information processing device, and may be, for example, an information processing device such as a personal computer, tablet terminal, or smartphone used by a user as a general consumer or worker, or if the user is an organization such as a company, it may be a server operated by the company or organization. The user terminal 10 is connected to an external information processing device (for example, a server 20 to which each service unit 11 connects) via a communication network such as the Internet. The user terminal 10 may also be an information processing device owned by the user.

[0020] (2-2) Service Unit 11 The service unit 11 is a system that serves as an interface for providing services such as data analysis, medical care, education, procurement, sales, and purchasing, and the information processing system 1 always uses multiple service units 11. When switching service units 11 as described later, the use of some of the service units 11 currently in use is stopped, and a service unit 11 that is not currently in use is used. Here, "in use" of a service unit 11 may include, for example, that the service unit 11 is running or immediately executable. The usage history of service units 11, etc. is recorded in the PDS unit 12. The service unit 11 may be application software for using a specific service. The service unit 11 may be implemented by executing a program stored on the user terminal 10. The service unit 11 may be downloaded and installed on the user terminal 10 from, for example, an app store. Alternatively, the service unit 11 may be implemented by executing a program published as a Web API that operates externally (for example, provided by a linked server 20) via a communication network such as the Internet. The service unit 11 may have a function to receive operations from the user (operation reception unit), a function to execute processing according to the received operation (processing unit), and a function to output the processing result (output unit).

[0021] A user may use the service unit 11 directly, or the service unit 11 may use other service units 11. Here, the use of other service units 11 by the service unit 11 may include the service unit 11 using other service units 11 on behalf of the user. Specifically, this may include the service unit 11 making the other service unit 11 available (for example, by exchanging a contract), or the service unit 11 inputting data (for example, the user's health check data) into the other service unit 11 (for example, an insurance-related service).

[0022] The content of the services provided by the Service Department 11 is not particularly limited, but as an example, specific examples of service functions are shown in Figure 2. In the table shown in Figure 2, the "Category" column shows specific examples of service categories provided by the Service Department 11, such as "Education," "Nursing Care," "Medical Care," "Administration," "Investment," "Maternal and Child Health," "Document Creation," and "Purchasing." In addition, the "Function" column in the table shown in Figure 2 shows examples of functions that are expected to be provided for each service category. In the "Log" column in the table shown in Figure 2, examples of log content generated by the use of the service are shown for each service category.

[0023] (2-3) AI Unit 13 The AI ​​unit 13 is a service unit that provides services that meet the user's needs by searching for and executing appropriate services in response to multimodal dialogue with the user. The AI ​​unit 13 may also provide advice to the user or act as a conversation partner. The AI ​​unit 13 may have a function to receive operations from the user (operation reception unit), a function to execute processing according to the received operations (processing unit), and a function to output processing results (output unit).

[0024] The AI ​​unit 13 may acquire and use data from an external data supply system 15. For example, if the AI ​​unit 13 is a large-scale language model (LLM), the data acquired from the data supply system 15 may be used for fine-tuning the AI ​​unit 13 or as part of the prompts to the AI ​​unit 13.

[0025] The AI ​​unit 13 has the function of performing multimodal dialogue with the user. The AI ​​unit 13 may have functions such as a dialogue agent, chatbot, and AI speaker (for example, natural language dialogue). The method of dialogue by the AI ​​unit 13 is not particularly limited, but for example, it may be a rule-based method that returns a predetermined response for each keyword or pattern in the user's input, or it may be a method that generates responses using LLM.

[0026] The AI ​​unit 13 may accept various forms of input from the user via the input device 112. Specifically, the AI ​​unit 13 may accept input such as voice spoken by the user, text entered by the user, or gestures and body language from the user. The AI ​​unit 13 may determine the user's needs by analyzing the input from the user obtained through dialogue with the user, and based on the determination result, may select a service unit 11 to use from among a plurality of service units 11, and then execute the selected service unit 11.

[0027] The AI ​​unit 13 may prioritize the use of a specific service unit 11 if there are multiple service units 11 with similar functions. The specific service unit 11 may be determined based on the priority set for each service unit 11. This priority may be set or periodically updated based on evaluations by the evaluation unit 14, or it may be set arbitrarily by the user. The AI ​​unit 13 may limit the number of service units 11 called in each dialogue session (for example, one, two, or three or more). It is preferable that the AI ​​unit 13 is a function within the user terminal 10, but it is not limited to this, and may operate outside the user terminal 10 as a Web API or SaaS.

[0028] Furthermore, the information processing system 1 may include multiple AI units (first AI unit, second AI unit, etc.). For example, a single user may use multiple AI units simultaneously. Each AI unit included in the multiple AI units may correspond to, for example, a service category. The category may be assigned by, for example, an app store providing the service unit 11, or it may be arbitrarily set by the user. By having the AI ​​unit 13 correspond to a service category, it becomes possible to provide the user with more detailed services for a specific category.

[0029] The AI ​​unit 13 may understand the user's needs through dialogue, referencing information on various products and services obtained from an external data supply system 15 and data stored in the PDS unit 12, and select and execute an appropriate service from the services provided by the service unit 11. The content of the data handled by the AI ​​unit 13 is not particularly limited, but may include a variety of things such as the user's basic attribute information, as well as interests, affiliations, friendships, behavioral history, and purchase and usage history of products and services. Specifically, it may include information from social networks, purchase history, preference information (running records, hiking records, etc.), medical history, and bank deposit and withdrawal history. If the user is a company, the AI ​​unit 13 may handle data such as personnel, procurement, sales, and business planning. Much of this data is logs from the service unit 11.

[0030] The AI ​​unit 13 may output questions and suggestions to the user via the output device 114 of the user terminal 10, which will be described later. The AI ​​unit 13 may acquire speech data pre-stored in the user terminal 10 and output it, or it may acquire a catalog of services from an external data supply system 15 via a communication network and suggest the use of a service from that catalog that suits the user. The speech output by the AI ​​unit 13 is not limited to narrowly defined questions that explicitly ask something, but broadly includes remarks, calls, nods, etc. that prompt some kind of response from the user. The manner in which the speech is output is not particularly limited, but for example, if the output device 114 is configured as a display device, any content such as the text or images of the speech may be displayed on the display device. Alternatively, if the output device 114 is configured as a voice output device, the speech may be output as voice from the output device 114. The AI ​​unit 13 may have the output device 114, which is a display device, display a personified character and output the speech in a manner as if the character were speaking. The AI ​​unit 13 may, for example, adaptively modify subsequent utterances based on a model generated using machine learning, according to the content of the user's response obtained.

[0031] If the AI ​​unit 13 is like an LLM, there is a risk that it may make inappropriate statements or perform inappropriate services to the user due to inappropriate prompts, etc. Therefore, text and images (that are not formal language data) that may contain inappropriate prompts may be excluded from the data used by the AI ​​unit 13 to generate statements or to perform services of other service units or both. The AI ​​unit 13 can acquire data from other service units 11, PDS unit 12, and data supply system 15, but by limiting the data acquired from less reliable sources to formal language data, for example, it is highly likely that inappropriate statements and services can be detected and prevented using a strictly defined inference method with respect to formal language.

[0032] (2-4) Evaluation Unit 14 The evaluation unit 14 evaluates the units to be evaluated (service unit 11, PDS unit 12, data supply system 15, and data analysis system 16). The evaluation unit 14 may be configured to evaluate the units to be evaluated by, for example, analyzing the data stored in the PDS unit 12, and to disclose the evaluation results to the user. The evaluation unit 14 may have a function to receive operations from the user (operation reception unit), a function to execute processing according to the received function (processing unit), and a function to output processing results (output unit).

[0033] In the information processing system 1, the evaluation unit 14 may be configured to analyze only the data stored in the PDS unit 12 of users who use or own the user terminal 10. In addition, in the information processing system 1, the evaluation unit 14 may operate on an external server or the like, like the other service units 11, outside of the user terminal 10.

[0034] The evaluation unit 14 may output each evaluation result to the output device 114. The evaluation unit 14 may also send the evaluation results of the service unit 11 and notifications based on the evaluation results to a server 20 or the like associated with the service unit 11 being evaluated.

[0035] The content of the notification based on the evaluation results is not particularly limited, but for example, it may be a suggestion for improvement for the evaluated part that has a low evaluation. In this case, the notification may include information indicating which data supply system (or the data supplied by it) should be used to improve the evaluated part, as well as information indicating the trend and changes from past evaluations. The evaluation unit 14 may, for example, stop using the evaluated part if it determines that the evaluation is below a predetermined threshold.

[0036] Figure 2 shows specific examples of factors to be considered when the evaluation unit 14 evaluates each type of service. The factors to be considered in the evaluation can be arbitrarily set according to the content of the service provided by the service unit 11, and may include, for example, the user's purchase history, financial assets, health status, academic performance, changes in job position, and work performance. The evaluation method of the service by the evaluation unit 14 is not particularly limited, but for example, the evaluation may be calculated by considering the cost incurred for the service (amount, time, and effort) and the outcome (quality of the product, profit / loss of investment, and academic performance). It is desirable to standardize the evaluation criteria. For example, in the case of health and medical services, criteria such as the mental health status table published by the World Health Organization (WHO) may be used.

[0037] The evaluation by the evaluation unit 14 may include multiple perspectives. These perspectives may be arbitrarily set by the user. They may also be set according to the category of the service unit 11. The evaluation unit 14 may perform a comparison of at least two of the multiple service units 11. The comparison method is not particularly limited, but for example, an evaluation may be calculated for the service units 11 to be compared, and based on that evaluation, superiority or inferiority or ranking may be assigned to the service units 11 to be compared. The evaluation unit 14 may arbitrarily select the service units 11 to be compared, but for example, it may select the service units 11 specified by the user. The evaluation unit 14 may make selections based on categories, evaluations, etc., associated with the service units 11. In particular, the evaluation unit 14 may select at least two service units 11 associated with the same or similar categories as comparison targets. Furthermore, the evaluation unit 14 may include these new service units 11 and existing service units 11 in its comparison by acquiring data from the data supply system 15 when new service units 11 are released from websites or app stores, or when rankings or other evaluations of existing services are published or updated. The evaluation unit 14 can then, for example, compare the service unit 11 that a user is currently using with these new service units 11 and existing service units 11.

[0038] The evaluation unit 14 may output evaluation results for a specific unit being evaluated, or recommendations to use several specific units being evaluated that have high evaluations. For example, the evaluation unit 14 may recommend that the user stop using a data supply system whose supplied catalog content is biased and instead use a data supply system with comprehensive catalog content. It may also recommend the use of a data analysis system for which the user pays a large fee for data supplied from the PDS unit 12, or a data analysis system that performs data analysis of great social significance. For example, if the user is a manufacturing company, the evaluation unit 14 may recommend the use of a service unit 11 that corresponds to the sales destinations or parts suppliers of its products. The evaluation unit 14 may also be evaluated by other evaluation units 14, just like other service units 11. Furthermore, the evaluation unit 14 may change its recommendation content based on the difference between the evaluation of a unit currently being evaluated and the evaluation of a unit not currently being evaluated. The evaluation unit 14 may strengthen the recommendation expression as the difference increases (for example, when it exceeds a predetermined threshold). Furthermore, the evaluation unit 14 may prioritize the services selected by the AI ​​unit 13 or forcibly set the services.

[0039] The recommendations output by the evaluation unit 14 may include, for example, a recommendation to integrate the service units 11. That is, the recommendations may include a recommendation to switch from multiple (for example, two) service units 11 to at least one other service unit 11 (for example, one) (stop using the former and use the latter). In this case, the recommendations may include information based on a comparison before and after the switch (such as differences in the functions of the service units 11 in the information processing system 1 before and after the switch). Alternatively, the recommendations output by the evaluation unit 14 may include a recommendation to disassemble the service units 11. That is, the recommendations may include, for example, a recommendation to switch from at least one (for example, one) service unit 11 to multiple other service units 11 (for example, two) (stop using the former and use the latter). In this case, the recommendation may include information based on a comparison before and after the switch (such as differences in the functionality of the service unit 11 within the information processing system 1 before and after the switch). The evaluation unit 14 may output the content and scope of logs necessary for the use of at least one of the service units 11 included in the recommendation. The evaluation unit 14 may perform evaluations of each data supply system 15, evaluations of each data supply system 15 in each category of service, evaluations of each combination of service unit 11 and data supply system 15, and recommend switching to a different data supply system 15 according to the evaluation results. For example, the user terminal 10 may output evaluation results from multiple evaluation units 14 for the same service unit 11 and accept a selection from these multiple evaluation results by the user. The user terminal 10 may then determine the selected evaluation result as the final evaluation result and use that evaluation result for various recommendations, etc.

[0040] The evaluation unit 14 may evaluate the reliability of the data sources used by the AI ​​unit 13 to generate utterances or to perform services of other service units, or both. For example, if information regarding the reliability of various service units 11 is obtained from a reliable data supply system 15, the evaluation unit 14 may use this to evaluate the reliability of the service units 11.

[0041] The evaluation of the evaluated unit by the evaluation unit 14 makes it easier for users to choose better services. As a result, market principles can lead to an overall improvement in the quality of services across society, and user trust in services can be fostered and maintained.

[0042] (2-5) PDS Unit 12 The PDS unit 12 has the function of storing various data related to the user and processing and outputting it as necessary. This data may include input / output logs of each service unit 11. The PDS unit 12 may store authentication credentials, etc., as an identity wallet. The PDS unit 12 may store data inside the user terminal 10, or it may store data in a predetermined storage unit provided outside the user terminal 10, or it may store data inside or outside the user terminal 10. The PDS unit 12 may store logs in association with the service unit 11 that generated the logs. The PDS unit 12 may have a function to receive user operations (operation reception unit), a function to execute processing according to the received operations (processing unit), and a function to output processing results (output unit).

[0043] The PDS unit 12 may store data so that only the user or a person authorized by the user can access it. The PDS unit 12 may grant access to the data when it is determined that the user has the right to access the data, for example, when a request for access to the data is made. The PDS unit 12 may store the data in an encrypted state. Here, encrypted state means making it difficult to access the data. For example, encrypted state includes encryption and secret sharing. In encryption, the original data (plaintext) is encrypted using a specific key, converted into ciphertext, and then transmitted. The recipient obtains the data by decrypting it. Encryption may include, for example, symmetric key cryptography and public key cryptography. Symmetric key cryptography is a method in which the sender and receiver use the same key to encrypt and decrypt the data, and may include, for example, AES (Advanced Encryption Standard). Public key cryptography is a method in which the sender and receiver use different keys, with encryption using the public key and decryption using the private key. Public-key cryptography may include, for example, RSA (Rivest-Shamir-Adleman). Secret sharing is a technique that divides data into multiple parts (shares) and stores each in a different location. The original data cannot be restored unless a predetermined number of shares are collected, and the confidentiality of the whole is maintained even if some shares are leaked. The data stored in the PDS unit 12 may include information such as the user's purchase history, financial assets, health status, academic performance, changes in job position, sales performance, and business plan. The PDS unit 12 can supply this data to the service unit 11 and the data analysis system 16. The PDS unit 12 can process the data it supplies to the service unit 11 and the data analysis system 16 as needed. This is an anonymization that allows the service unit 11 and the data analysis system 16 to perform processing with the required accuracy without passing any extraneous information to them, and may use differential privacy, cryptographic computation, multi-party computation, or federative learning. For example, if the user is a business that manufactures machine parts, the PDS unit 12 can supply detailed specification data of the parts to the service unit 11, which is responsible for procurement at the companies to which the parts are delivered, without disclosing the details of the parts' manufacturing method.Similarly, it is possible to supply the service unit 11 in charge of procuring the component materials with the data of the required specifications of the desired materials without disclosing the detailed specifications of the components.

[0044] (2-6) Data Supply System 15 The data supply system 15 is, for example, outside the information processing system 1, and there may be one or a plurality of them. The data supply system 15 supplies data including comprehensive knowledge (including information on products and services) to the service unit 11. The data includes the catalog of the service unit 11 (including information on products and services; the same applies hereinafter), the catalog of the PDS unit 12, the catalog of the data supply system 15, the catalog of the data analysis system 16, medical knowledge for the general public, a glossary of terms in a certain specialized field, and a collection of papers in a certain field. The catalog of the service unit 11 includes a catalog of administrative services for the residents of a certain municipality, a catalog of automobiles, rental houses, and cosmetics, a catalog of nursing facilities and medical institutions in a certain area, and a catalog of businesses that procure certain types of components. For example, a certain service unit 11 obtains a catalog of municipal resident services from the data supply system 15, selects a service that matches the user data stored in the PDS unit 12 and the user's utterance from the catalog, and calls another service unit 11 that provides the service to execute the service. The data obtained from the data supply system 15 may be used, for example, for fine-tuning the AI unit 13 included in the service unit 11 or included in the prompt to the AI unit 13.

[0045] (2-7) Data Analysis System 16 The data analysis system 16 is located outside the information processing system 1 and may be one or multiple. The data analysis system 16 is a system that analyzes the data for some purpose, such as the development and improvement of the information processing system 1, the PDS unit 12, and the service unit 11, the creation, augmentation, and correction of data supplied by the data supply system 15, research and development of products and services, academic research, and policy formulation and evaluation. In other words, the developers of the information processing system 1, the PDS unit 12, and the service unit 11, the operators of the data supply system 15, the providers of products and services, research institutions, governments and local authorities, etc., operate the data analysis system 16. For example, the operator of the data supply system 15 that supplies a catalog of cosmetics may use the data analysis system 16 to predict the compatibility between various cosmetics and individuals with diverse attributes, and supply the catalog with the prediction results included. Thus, the information such as catalogs supplied by the data supply system 15 may include information such as the compatibility between products and services and user attributes, which is the result of calculation (prediction) using the data analysis system 16.

[0046] (3) Hardware Configuration Diagram 3 is a block diagram showing an example of the schematic configuration of the hardware of the user terminal 10 according to the embodiment. Note that the server 20 may also have a hardware configuration similar to that of the user terminal 10 described below.

[0047] As shown in Figure 3, the user terminal 10 includes, for example, a CPU (Central Processing Unit) 102, RAM (Random Access Memory) 104, ROM (Read Only Memory) 106, a drive device 108, a network interface 110, an input device 112, and an output device 114. These components are connected to each other via a bus, enabling data transmission and reception.

[0048] The CPU 102 is an example of a processor that controls each device and performs data calculation and processing. The CPU 102 executes programs stored in a memory such as the RAM 104 or the ROM 106. The CPU 102 receives data from the input device 112, the network I / F 10, etc., performs calculation and processing, and then outputs the calculation result to the output device 114, etc.

[0049] The RAM 104 is, for example, a memory configured as a main storage unit. The RAM 104 is a storage device that stores programs and data such as an OS (Operating System), which is basic software executed by the CPU 102, and application software.

[0050] The ROM 106 is, for example, a memory configured as an auxiliary storage unit, and is a memory that stores data related to application software, etc.

[0051] The drive device 108 reads programs and data from a recording medium 116, such as a CD-ROM or an SD card, and installs them in the memory.

[0052] Further, the recording medium 116 stores predetermined programs and data, and the programs and data stored in this recording medium 116 are installed in the user terminal 10 via the drive device 108. The installed predetermined program can be executed by the user terminal 10.

[0053] The network I / F 110 is an interface between the user terminal 10 and a peripheral device having a communication function. Also, the network I / F 110 is connected via a network such as a LAN (Local Area Network) or a WAN (Wide Area Network) constructed by a data transmission path such as a wired and / or wireless line.

[0054] The input device 112 may include, for example, a keyboard equipped with cursor keys, numeric inputs, and various function keys, or a mouse or slide pad for selecting keys on the display screen of the output device 114, which is configured as a display device. The input device 112 may also be configured as an audio input device, such as a microphone capable of receiving user voice input. The input device 112 may also be configured as an imaging device, such as a camera. Furthermore, the input device 112 is a user interface for the user to give operation instructions to the CPU 102 or input data.

[0055] The output device 114 is configured, for example, as a display device and / or an audio output device. The output device 114 may be configured as a display device using, for example, an LCD (Liquid Crystal Display), and may display information according to the display data input from the CPU 102. Alternatively, the output device 114 may be configured as an audio output device using a speaker, and may output audio according to the audio data input from the CPU 102. The input device 112 and the output device 114 may be provided outside the user terminal 10.

[0056] (4) Operation process diagram 4 is a flowchart showing an example of operation processing by the information processing system 1 according to the embodiment.

[0057] (S201) First, input / output of the service unit 11 is performed in conjunction with the use of the service provided by the service unit 11. For example, the input / output may be a dialogue between the user and the service unit 11. Alternatively, the input / output may be between service units 11. Alternatively, the input / output may be the supply of data from the PDS unit 12 or the data supply system 15 to the service unit 11. The user may use other service units 11 through one service unit 11. The AI ​​unit 13 may perform a multimodal dialogue with the user and search for and execute an appropriate service unit 11 according to the dialogue. The evaluation unit 14 may be used for this search. The AI ​​unit 13 may also be the evaluation unit 14.

[0058] Here, an example of interaction between the user and the AI ​​unit 13 when the user uses the service unit 11 via the AI ​​unit 13 is shown below. As mentioned above, the AI ​​unit 13 can also accept input via the user's voice and gestures, but the following example shows the transcribed text of the voice dialogue between the user and the AI ​​unit 13.

[0059] <Example 1: Subscription Service Example> User: I don't use this subscription much, so I'd like to cancel it. AI Department: Okay, your subscription has been canceled.

[0060] In Example 1, for example, if the AI ​​unit 13 receives input from a user indicating that they "want to cancel their subscription," it will communicate with the service unit 11, which manages the subscription, and execute a process to cancel the subscription. For example, the service unit 11, having received instructions from the AI ​​unit 13, will obtain the user's personal information and subscription-related data from the PDS unit 12, and then send a cancellation request containing this information to the relevant server 20. The AI ​​unit 13 may also ask the user which subscription it is. The AI ​​unit 13 may also notify the user of the information indicating the cancellation result obtained by the service unit 11 from the server 20.

[0061] <Example 2: Example of administrative services> AI Department: Your income has decreased, so you are eligible for child allowance from the city. I will apply for it. User: Okay, please do. AI Department: I have applied. You should receive 25,000 yen starting next month.

[0062] In Example 2, the AI ​​unit 13 obtains data on the user's income via the PDS unit 12 and then determines whether the user is eligible for child allowance based on their income. The data on the user's income may be stored in the PDS unit 12 as the user's personal information or trade secrets, or it may be logs stored in the PDS unit 12 associated with the service unit 11 of a predetermined financial service, administrative service, or any other service. Information on eligibility for child allowance may be obtained from a catalog of administrative services of the city hall obtained from the data supply system 15. If the AI ​​unit 13 determines that the user is eligible based on their income, it proposes that the user apply for child allowance. In this way, the AI ​​unit 13 may obtain data stored in the PDS unit 12 and make various judgments. If, as a result of its judgment, the AI ​​unit 13 determines that a certain service would be beneficial to the user (not limited to monetary benefits, but broadly including benefits from a health or other perspective), it may notify the user of a recommendation for that service. When the AI ​​unit 13 understands the user's input "Okay, please proceed" as permission, it communicates with the administrative service unit 11 and executes the process for applying for child allowance. Through this process, the administrative service unit 11 sends the child allowance application data to an external server 20 (such as a server for a system managed by the city hall). The AI ​​unit 13 may also notify the user of the application results and the status of child allowance acquisition obtained by the service unit 11 from the server 20.

[0063] <Example 3: Examples of business services and scheduling services> - User: Please apply for paid leave the day after tomorrow. - AI Department: You will be on a business trip to Sendai starting the day after tomorrow.

[0064] In Example 3, if the user requests "Please submit a paid leave request for the day after tomorrow," the AI ​​unit 13 obtains information on events already scheduled for that day from the PDS unit 12, etc., and then checks the user's schedule. This data may be stored by the PDS unit 12 as the user's personal information or trade secrets, or it may be a log stored in association with a service unit 11 of a predetermined schedule service, business support service, or any other service. The AI ​​unit 13 confirms from the schedule data that the user has a business trip to Sendai scheduled for the day after tomorrow, and if it determines that the paid leave request is inappropriate, it notifies the user accordingly. In this way, when schedules conflict, the AI ​​unit 13 may determine the importance of the conflict using various indicators and decide whether the user's request is appropriate or not. The AI ​​unit 13 may also suggest schedule changes to the user (alternative dates for submitting the paid leave request or alternative dates for the business trip), or, if it receives instructions to change the schedule, it may change the schedule via the service unit 11 of the schedule service.

[0065] <Example 4: Example of an e-commerce service> User: My shoes are worn out, so I'd like to buy new ones. AI: You don't wear leather shoes very often, so it's the sneakers you bought two years ago, right? User: That's right. AI: We don't sell the same model anymore, so how about these instead? (Shows two photos) The size is right. User: The black ones look better. AI: We have them in stock at our warehouse in Shinagawa, but a suitable truck won't be available for three days. User: I'd like them tomorrow if possible. AI: Then it will cost 200 yen more, but we'll arrange for them to arrive tomorrow. We paid 5,200 yen to the manufacturer and 300 yen to the shipping company.

[0066] In Example 4, the AI ​​unit 13 obtains data on the user's purchase history from the PDS unit 12, etc., and then checks information on shoes previously purchased by the user. If the user requests "buy me a new pair," the AI ​​unit 13 identifies the type and size of shoes previously purchased by the user based on the previous purchase history. The AI ​​unit 13 obtains the inventory status of the previously purchased shoes from the service unit 11 that provides e-commerce services, and if the same type of shoes are not available, it suggests alternative products to the user by displaying alternative options. If the user selects an alternative product, the AI ​​unit 13 similarly obtains the inventory location from the same service unit 11 and presents a delivery schedule using another service unit 11 that provides logistics services. Depending on the user's wishes, the AI ​​unit 13 calculates additional charges and requests the service unit 11 to arrange for prompt delivery. In the delivery arrangement process, the AI ​​unit 13 may allow only the logistics company performing the delivery to view the user's personal information, such as their address and name.

[0067] <Example 5: Example of tax services> ・AI Department: It's March, so shall we file your tax return? ・User: Oh, yes, please. ・AI Department: Do you have a paper withholding tax statement? You gave a lecture at University A, right? ・User: Ah, yes, I do. ・AI Department: I'll scan it, so please hold it here. ・User: Okay. ・AI Department: Your lecture fee was 110,000 yen and the withholding tax amount is 11,231 yen. You bought research materials from a consultant for 77,000 yen for the lecture, right? Is it okay to consider that a necessary expense? ・User: Yes. ・AI Department: (Displays tax return) This is your tax return. Your income has increased by about 800,000 yen since last time. You have made estimated tax payments, so 150,200 yen will be deducted. Is it okay to use Bank B for the deduction account? ・User: No, please use Bank C. ・AI Department: Okay, the deduction will be transferred to Bank C. ・User: Thank you very much.

[0068] In Example 5, the AI ​​unit 13 obtains data on the user's income from the PDS unit 12, etc., and then confirms the need for filing a tax return. If the user requests "Please do so," the AI ​​unit 13 begins the tax return process. First, the AI ​​unit 13 checks information such as the user's withholding tax slips and past lecture records, and asks the user to scan the withholding tax slips. Once the user scans the withholding tax slips, the AI ​​unit 13 obtains the data and performs the necessary calculations. At this time, the user may use a document creation app or scanning app on their smartphone. Next, the AI ​​unit 13 creates a tax return based on the obtained data and calculates necessary expenses and the amount of tax payable. After that, the AI ​​unit 13 has the user confirm the tax return and specify the bank account for the payment. If the user selects Bank C, the AI ​​unit 13 specifies the payment destination accordingly. Finally, the AI ​​unit 13 notifies the user that the contents of the tax return and the payment transfer procedure are complete.

[0069] (S202) Next, the service log generated by the input / output in step S201 is stored in the PDS unit 12. Specifically, for example, the AI ​​unit 13 may acquire the log and store it in the PDS unit 12. Alternatively, the service unit 11 may store the log in the PDS unit 12. The PDS unit 12 may store the log so that only the user or a person authorized by the user can access it. Furthermore, the PDS unit 12 may store a log that includes information about what input / output (for example, from the user to a specific service unit 11) was used to generate the log.

[0070] Steps S201 and S202 described above may be executed each time a service provided by the various service units 11 is used. In this way, various logs related to the user's use of the various services are accumulated in the PDS unit 12.

[0071] (S203) The evaluation unit 14 evaluates the units to be evaluated, including those not currently in use, by analyzing the data stored in the PDS unit 12, and discloses the evaluation results to the user. It may also disclose several units to be evaluated that have high evaluations and their evaluation values. The timing at which this process is executed is not particularly limited, but may be, for example, when the user makes a request to the evaluation unit 14 directly or via the AI ​​unit 13, or at predetermined periodic or irregular timings.

[0072] (S204) The evaluation unit 14 outputs information regarding the evaluation results (evaluation information).

[0073] Figure 5 is a diagram showing an example of an evaluation information screen according to an embodiment. The screen 300 shown in Figure 5 is an example of a screen when an evaluation is performed on two services, Service A and Service B, which belong to a specific category. This screen is displayed on a display device such as a user terminal 10. Here, Service A is a service that the user is currently using, and Service B is a service that belongs to the same category as Service A but is not currently being used by the user. The screen 300 includes a display area 301 and a selection area 302.

[0074] The display area 301 contains scores for various aspects of Service A and Service B in a table format. Specifically, the display area 301 shows the overall score for both services and the scores for various aspects (aspect a, aspect b, ...). These scores may be calculated by the evaluation unit 14. The higher score for Service A and Service B may be highlighted. In the example shown in Figure 5, the higher score for Service A and Service B is bolded and underlined. This makes it easy for the user to compare services not only overall but also services from different perspectives.

[0075] The evaluation unit 14 may, as a result of evaluating the services, display a selection unit for the user to choose to switch services if the evaluation of a service unit 11 other than the currently used service unit 11 is higher than the evaluation of the currently used service unit 11. In Figure 5, an example of a selection unit is shown, a selection unit 302 with the text "Switch to Service B". The PDS unit 12 may execute a process to switch services in response to the selection unit 302 being selected. In the example in Figure 5, for example, if the user selects the selection unit 302, the PDS unit 12 executes a process to switch from Service A to Service B (stop using the service unit 11 corresponding to Service A and start using the service unit 11 corresponding to Service B). As a result, the PDS unit 12 may change its settings to switch the service to be used preferentially from Service A to Service B in response to interaction with the user.

[0076] Furthermore, the PDS unit 12 may supply data to service B when the selection unit 302 is selected and service B is to be used. This allows the service unit 11 that provides service B and the server 20 that cooperates with the service unit 11 to use the data.

[0077] In this way, by disclosing the evaluation of the evaluated part to the user, the user can choose a better evaluated part, and through market principles, the quality of services provided by the information processing system 1 will improve for society as a whole, and the trust of many users in the system can be fostered and maintained.

[0078] 1 Information Processing System 10 User Terminal 11 Service Unit 12 PDS (Personal Data Store) Unit 13 AI Unit 14 Evaluation Unit 15 Data Supply System 16 Data Analysis System 20 Server 108 Drive Device 112 Input Device 114 Output Device 116 Recording Medium 300 Screen 301 Display Area 302 Selection Unit

Claims

1. An information processing method performed by at least one computer, comprising: a plurality of service units inputting and outputting data to each other; a dialogue unit included in the plurality of service units interacting with a user; storing logs of the input and output of the plurality of service units in a PDS unit; and processing the data stored in the PDS unit as needed and supplying it to the plurality of service units.

2. The dialogue unit according to claim 1, wherein all or part of the data other than formal language data is removed from the data used to generate utterances or to perform other service units or both.

3. An evaluation unit that evaluates at least one of the plurality of service units and PDS units, and uses the result of the evaluation to select the one with the higher evaluation among the plurality of service units and PDS units, and is included in the plurality of service units described in claim 1.

4. An information processing device comprising: a plurality of service units that input and output data to and from each other; a dialogue unit included in the plurality of service units that interacts with a user; and a PDS unit that stores logs of input and output from the plurality of service units, processes the stored data as needed, and supplies it to the plurality of service units.

5. A program for causing at least one computer, which comprises multiple service units and a PDS unit, to execute the following steps: the multiple service units input and output data to and from each other; a dialogue unit included in the multiple service units interacts with a user; a log of the input and output of the multiple service units is stored in the PDS unit; and the data stored in the PDS unit is processed as necessary and supplied to the multiple service units.

Images