Information processing device, information processing method, and program

Abstract

To provide an information processing device, an information processing method, and a program that can interpret user requests and select an appropriate response channel. [Solution] An information processing device comprising: a first processing unit that acquires text information which is a text representation of a request issued by a user, and inputs input information including the text information into a first model to acquire first model output information which represents the request output by the first model; and a second processing unit that inputs at least the first model output information into a second model which is constructed as a model different from the first model to acquire second model output information which indicates which of a plurality of corresponding channels output by the second model matches the user's request, and selects a corresponding channel which corresponds to the user's request based on the second model output information.

Description

Technical Field

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

Background Art

[0002] Conventionally, in various services, attempts have been made to automatically give some answer to natural language input in the form of voice or text. For example, Patent Document 1 discloses an invention in which a generative AI is used to make an abstract question from a user more specific, and an answering system presents an answer sentence based on the keywords output by the generative AI. Further, Patent Document 2 discloses an invention in which comment sentences including question keywords included in a question sentence are searched from a database, they are classified into a plurality of topic groups from the distribution of appearance words, and the similarity between the group of comment sentences included in each topic group and the sentences included in each answer sentence is calculated, and an answer sentence with a high similarity is output.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventional techniques only output an answer sentence in some form, and there are cases where the user's request cannot be interpreted and an appropriate response channel cannot be selected.

[0005] The present invention has been made in consideration of such circumstances, and one of its objects is to provide an information processing apparatus, an information processing method, and a program capable of interpreting a user's request and selecting an appropriate response channel.

Means for Solving the Problems

[0006] One aspect of the present invention is an information processing device comprising: a first processing unit that acquires text information which is a text representation of a request issued by a user, and inputs input information including the text information into a first model to acquire first model output information which represents the request output by the first model; and a second processing unit that inputs at least the first model output information into a second model which is constructed as a model different from the first model to acquire second model output information which indicates which of a plurality of corresponding channels matches the user's request output by the second model, and selects a corresponding channel which corresponds to the user's request based on the second model output information. [Effects of the Invention]

[0007] According to one aspect of the present invention, it is possible to interpret the user's request and select an appropriate response channel. [Brief explanation of the drawing]

[0008] [Figure 1] This diagram shows the basic forms of in-store electronic payment. [Figure 2] This diagram shows an example of a configuration for performing electronic payments (terminal payments) using a payment app. [Figure 3] This figure shows an example of the contents of user information 191. [Figure 4] This figure shows an example of the contents of merchant / store information 192. [Figure 5] This diagram shows an overview of the processing flow when a user scan is performed. [Figure 6] This diagram shows an overview of the processing flow when a store scan is performed. [Figure 7] This diagram shows an example of a configuration for performing electronic payments (card payments) using payment cards. [Figure 8] This diagram shows an overview of the processing performed by the information processing device. [Figure 9]This diagram shows an overview of the processing performed by the estimation unit 150. [Figure 10] This is a diagram (part 1) illustrating the operation of an information processing device. [Figure 11] This is a diagram (part 2) illustrating the operation of an information processing device. [Modes for carrying out the invention]

[0009] [overview] The following describes embodiments of the information processing device, information processing method, and program according to the present invention with reference to the drawings. The information processing device provides additional functions to network services, such as electronic payment services. Network services have response channels such as frequently asked questions, chatbots, automated procedures, coupon issuance, and human support. The information processing device acquires text information that is text-based from requests made by users in natural language form, such as voice or text input, and selects a response channel through the cooperation of a first model and a second model configured separately from each other. The information processing device may be a function of a service server of a network service (a payment server in the case of an electronic payment service), i.e., a virtual machine, or it may be a device configured independently of those servers. In the following description, the information processing device is assumed to be a function of a payment server that provides electronic payment services.

[0010] This document describes electronic payment services. An application program, a payment server, and a credit card server work together to provide electronic payment services. In the following description, the application program will be referred to as the payment app. Electronic payment services are services that support payments for the purchase of goods and services at stores. A store refers to, for example, a physical store (a brick-and-mortar store) that exists in the real world, but may also include a virtual store for e-commerce. A virtual store may include one provided by an entity different from the operator of the electronic payment service. In that case, when settling a purchase at a virtual store, the user is controlled to be redirected to the electronic payment service interface screen. In electronic payment services, stores are treated as belonging to, for example, affiliated merchants (brands), and processing such as payment when a purchase is made at a store is mainly carried out between the user and the affiliated merchant. Alternatively, processing such as payment may be carried out between the user and the store.

[0011] The payment app is equipped with a function to accept requests via voice or text input. The information processing device, for example, retrieves the requests received by the payment app and performs the above processing. Alternatively, the information processing device may obtain requests via the electronic payment service's website.

[0012] [Types of in-store electronic payment methods] Figure 1 shows a basic configuration of in-store electronic payment. Basically, electronic payment is executed by three parties: a medium M held by the user U, store equipment E, and a payment system S. The medium M is a portable computer device such as a smartphone or a credit card. Store equipment E is located in a physical store (hereinafter simply referred to as "store") in the real world and includes POS devices, wireless communication devices, credit card readers, printed materials with code images such as QR codes (registered trademarks), or display devices that show code images. In in-store electronic payment, first, user identification information and payment amount information are shared unidirectionally or bidirectionally between the medium M and the store equipment E. During this process, one of the medium M or store equipment E optically reads various information from a code image displayed by the other, provides information via NFC (Near Field Communication), or reads the PAN (Primary Account Number) by a credit card reader. Then, one of the medium M or store equipment E (the one that received information from the other) transmits the payment information necessary for payment to the payment system S via the network NW. Furthermore, both the medium M and the store equipment E may transmit some information to the payment system S. The payment system S manages various information of the user U and performs electronic payments between the store and the user U in various ways. Electronic payments are made using either a prepaid system or a post-paid system, or both, or by other methods. In addition, electronic payments may also include forms of so-called online shopping, which are performed by both the user's terminal device and the payment system. The network NW includes, for example, the internet, LAN (Local Area Network), wireless base stations, and provider equipment. Various devices that communicate via the network NW, as described later, are assumed to have communication devices such as network cards and wireless communication modules.

[0013] [Configuration (Terminal Payment)] FIG. 2 is a diagram showing an example of a configuration for performing electronic payment (terminal payment) using a payment application. This electronic payment is executed centering around a payment application 20 operating on a user terminal device 10 which is one of the media M, one or more store payment terminals 30 which are one of the store facilities E, one or more store code images 40, and a payment server 100 which constitutes part of a payment system S. The payment server 100 communicates with the user terminal device 10, the store payment terminal 30, and one or more information terminals 50 via a network NW.

[0014] The user terminal device 10 is, for example, a portable terminal device such as a smartphone or a tablet terminal. The user terminal device 10 is a computer device having at least an optical reading function, a communication function, a display function, an input reception function, and a program execution function. In the following description, the configurations for realizing these functions are respectively referred to as a camera, a communication device, a touch panel, a CPU (Central Processing Unit), etc. In the user terminal device 10, the payment application 20 is executed by a processor such as a CPU, and operates to provide an electronic payment service to the user in cooperation with the payment server 100. The payment application 20 is installed in the user terminal device 10 from, for example, an application distribution server (not shown), and controls the camera, the communication device, the touch panel, etc. of the user terminal device 10. In the following description, cases where it is described as "transmitting information to the user terminal device 10 (or receiving / obtaining information from the user terminal device 10)" and cases where it is described as "transmitting information to the payment application 20 (or receiving / obtaining information from the payment application 20)" may be mixed, but these are only differences in expression and are not intended to distinguish anything.

[0015] The store payment terminal 30 is installed in a store, for example. The store payment terminal 30 is a computer device (or an aggregate thereof) having at least a product price acquisition function, an optical reading function, a program execution function, and a communication function. The store payment terminal 30 includes a so-called POS (Point of Sale) device, and the POS device may have a product price acquisition function or an optical reading function.

[0016] The store code image 40 is placed in the store and is a code image such as a QR code (registered trademark) printed on a paper or plastic medium. Note that the store code image 40 may be displayed by a display placed in the store (which may be a display of a terminal device such as a smartphone or a tablet terminal).

[0017] The information terminal 50 is used by the operator of a franchise that oversees the store. In an electronic payment service, a customer as a provider of goods or services is treated as a franchise (brand), and one or more stores exist under its umbrella. There may be a franchise that operates only one store. The information terminal 50 is a smartphone, a tablet terminal, a personal computer, or the like. The information terminal 50 operates a franchise interface 55. The franchise interface 55 may be a franchise application or a web page displayed by a general-purpose browser. The franchise interface 55 accepts settings of coupons by the operator of the franchise and transmits them to the payment server 100. The information terminal 50 may have a function of displaying a code image corresponding to the store code image 40 or reading a code image displayed by the user terminal device 10 by executing the franchise interface 55 (in the latter case, an optical reading function is required).

[0018] The payment server 100 communicates with the credit card server 200 via the network NW. The payment server 100 has, for example, a content providing unit 110, an information management unit 120, a payment processing unit 130, a first processing unit 140, an estimation unit 150, a second processing unit 160, a correspondence processing unit 170, and a storage unit 190. Components other than the storage unit 190 are realized, for example, by a hardware processor such as a CPU executing a program (software). Some or all of these components are LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable The program may be implemented by hardware (including circuitry) such as a Gate Array or a GPU (Graphics Processing Unit), or by the collaboration of software and hardware. The program may be stored in advance on a storage device such as an HDD (Hard Disk Drive) or flash memory (a storage device equipped with a non-transient storage medium), or it may be stored on a removable storage medium such as a DVD or CD-ROM (a non-transient storage medium) and installed on the storage device when the storage medium is inserted into a drive device.

[0019] The storage unit 190 can be an HDD, flash memory, RAM (Random Access Memory), etc. The storage unit 190 may also be a NAS (Network Attached Storage) device accessible by the payment server 100 via a network. The storage unit 190 stores information such as user information 191, merchant / store information 192, knowledge database 193, first model 194, and second model 195. The first processing unit 140, estimation unit 150, second processing unit 160, response processing unit 170, knowledge database 193, first model 194, and second model 195 combined constitute an example of an information processing device. These functions will be explained after the description of the electronic payment service.

[0020] The content provider unit 110, for example, has the functionality of a web server and provides information (content) for displaying various screens of the electronic payment service to the user terminal device 10. The content provider unit 110 provides content to the user terminal device 10 in the form of a web page, or provides the user terminal device 10 with parameters necessary for the payment application 20 to render images.

[0021] The Information Management Department 120 edits, adds, and deletes user information 191 and merchant / store information 192, and manages these.

[0022] Figure 3 shows an example of the contents of User Information 191. User Information 191 is a collection of information such as User URL, Account ID, Phone Number, Password, Registration Date, Charge Balance, Electronic Money Type, Terminal Payment Method, Card Payment Method, Various History Information, Identity Verification Flag, Name, Address, Date of Birth, Email Address, Bank Account, Postpay Settings, and Postpay Conditions Information, all of which are linked to each other. Hereafter, the instance of a user (electronic payment account) to which this information is linked may be referred to as an account. In the figure, items indicated by "-" indicate that they are not set.

[0023] The user URL is used for transferring funds between users. When registering for the electronic payment service, registration of a phone number and password is required. The account ID is issued to the user by the payment server 100. The registration date is the date the user registered for the electronic payment service (the date the account was created). The charge balance is information indicating the balance of electronic money set by the user beforehand by sending money to the account. Methods of transfer include depositing money into an ATM (Automatic Teller Machine) of a designated company (bank) and transferring money from a registered bank account. The type of electronic money is information indicating, for example, whether the electronic money can be withdrawn or can only be used for electronic payments. The terminal payment method is setting information indicating whether the user will make an electronic payment using the charge balance (balance payment) or a deferred payment in terminal payments. The card payment method is setting information indicating whether the user will make an electronic payment using the charge balance (balance payment) or a deferred payment in card payments. Various types of transaction history information include charge history, which shows the history of users sending money to electronic payment services in advance to increase their balance; payment history, which shows the details of each transaction made by the user (date and time, store ID of the store where the purchase was made, merchant ID, payment amount, payment method, etc.); P2P transfer history; P2P deposit history; and withdrawal history.

[0024] The "Verified" flag indicates whether the user has completed identity verification using an identification document. Post-payment becomes available only after identity verification is complete. In the diagram, the user with account ID "002" has not completed identity verification and therefore can only select balance payment as their terminal payment method. The bank account is the account number of a bank account into which funds can be deposited for the electronic payment service. The "Post-payment Settings" indicates whether the user has completed the necessary setup to enable post-payment. The "Post-payment Conditions" information shows various conditions for post-payment, such as the limit and the current month's usage amount.

[0025] User information 191 may include, in addition to the information shown in Figure 3, attribute information such as age and occupation, merchant information such as the type of business and payment location of the merchant used, whether or not bank loans are used, whether or not there are any delinquencies in card loans, asset management information such as whether or not an account is held, investment experience, and motivation for use, coupon information showing details of coupon acquisition and use, and electronic flyer information such as the type of business of the merchant related to the viewed electronic flyer and payments made after the electronic flyer was viewed.

[0026] Figure 4 shows an example of the contents of the merchant / store information 192. The merchant / store information 192 includes, for example, a first table 192A in which merchant IDs and store IDs are associated with store URLs, a second table 192B in which merchant names and sales figures (as described above) are associated with merchant IDs, and a third table 192C in which store IDs are associated with store IDs. In addition to this information, the merchant / store information 192 may also include information such as the merchant or store category, store location, and payment patterns.

[0027] The payment processing unit 130 performs various processes for electronic payment. There are two methods for terminal payment, which are described below: the first method (user scan) and the second method (store scan).

[0028] Figure 5 shows an overview of the processing flow when a user scan is performed. First, the user terminal device 10, with the payment application 20 running, reads and decodes the store code image 40 using its optical reading function (S1). The store code image 40 contains information about the store URL. The payment application 20 sends first payment information, including the store URL and the user's account ID, to the payment server 100 (S2). The payment server 100 searches for merchant / store information 192 using the merchant ID and store ID corresponding to the store URL, obtains the merchant name and store name information (S3), and sends it to the payment application 20 (S4). The user enters the payment amount into the payment application 20 on the screen where the merchant name and store name are displayed (S5). Then, the payment application 20 generates second payment information, including at least the payment amount, and sends it to the payment server 100 (S6).

[0029] The payment processing unit 130 of the payment server 100 performs electronic payment based on the received second payment information if the "terminal payment method" in the user information 191 of the user is set to "balance payment" (S7-1). At this time, the payment processing unit 130 performs electronic payment by, for example, decreasing the charge balance managed in association with the user ID and increasing the item value of the merchant's sales proceeds. The item value of the merchant's sales proceeds is not used as electronic money itself, for example, but the amount corresponding to the item value of the sales proceeds is transferred to the bank account in a cycle according to the agreement between the merchant and the electronic payment service. On the other hand, if the "terminal payment method" is set to "post-payment", the payment processing unit 130 sends the first payment information and the second payment information to the credit card server 200 to request electronic payment (S7-2). The credit card server 200 performs electronic payment by adding the payment amount to the user's monthly usage amount based on the received information and deducting the monthly usage amount from the user's bank account after the closing date (S7-3).

[0030] Then, the payment processing unit 130 sends a payment completion notification (information for displaying the payment completion screen) to the payment application 20 via the content provision unit 110 (S8), and the payment application 20 displays the payment completion screen (S9). If the store code image 40 is displayed on a display placed in the store, the store code image 40 may include payment amount information as well as the store URL. In this case, the procedure for the user to enter the payment amount is omitted, and the payment amount information is included in the first payment information and sent to the payment server 100. Merchant name and store name information may be included and displayed on the payment completion screen.

[0031] Figure 6 shows an overview of the processing flow when a store scan is performed. First, when the payment app 20 is launched, when a payment operation is performed in the payment app 20, when it is time for an automatic update (for example, every minute), and at other times, the payment app 20 sends a request to the payment server 100 to issue a one-time code (S11). The payment processing unit 130 of the payment server 100 generates a one-time code (S12) and sends it to the payment app 20 (S13). The payment app 20 displays a code image such as a QR code or barcode that was generated based on the one-time code (S14). The user holds the display surface of the user terminal device 10 over the store payment terminal 30, and the store payment terminal 30 reads and decodes the code image using its optical reading function and obtains the one-time code, etc. (S15). Then, the store payment terminal 30 generates payment information including the one-time code, payment amount, merchant ID, store ID, etc., and sends it to the payment server 100 (S16). Payment amount information is obtained in advance through methods such as barcode scanning or manual entry.

[0032] The payment processing unit 130 of the payment server 100 identifies the user corresponding to the one-time code based on the received information, and if the "terminal payment method" in the user information 191 of that user is set to "balance payment", it performs electronic payment based on the received second payment information (S17-1). The content of the processing at this time is the same as the processing in S7-1 in Figure 5. On the other hand, if the "terminal payment method" is set to "post-payment", the payment server 100 sends the first payment information and the second payment information to the credit card server 200 to request electronic payment (S17-2). The credit card server 200 adds the payment amount to the user's monthly usage amount based on the received information, and performs electronic payment by deducting the monthly usage amount from the user's bank account after the closing date (S17-3).

[0033] Then, the payment processing unit 130 sends a payment completion notification to the payment application 20 via the content provision unit 110 (S18), and the payment application 20 displays a payment completion screen (S19).

[0034] Furthermore, electronic payment may be performed using only one of the above patterns. Also, the "account ID" explained in Figure 2 may be other information that can be used as user identification information (for example, a phone number). In addition, the issuance of a one-time code may be omitted during store scanning, and the payment app 20 may display a code image generated based on the user's account ID. In that case, the payment server 100 will identify the user corresponding to the account ID instead of identifying the user corresponding to the one-time code.

[0035] Furthermore, instead of managing the "post-payment" settlement through the credit card server 200, it may be handled internally by the payment server 100. In this case, the configuration of the payment card 60, credit card server 200, etc., may be omitted.

[0036] [Configuration (Card Payment)] Figure 7 shows an example of a configuration for electronic payment (card payment) using a payment card. This electronic payment is executed around a payment card 60, which is one of the media Ms; a credit processing terminal 70, which is one of the store equipment Es; and a payment server 100 and a credit card server 200, which constitute part of the payment system S. The credit card server 200 communicates with the credit processing terminal 70 via a network NW.

[0037] The credit processing terminal 70 is installed in the store, similar to the store payment terminal 30. The credit processing terminal 70 includes, for example, a credit payment terminal (credit card reader) and a POS device. The credit payment terminal reads the PIN (Personal Identification Number) from the inserted or scanned credit card and verifies it against the PIN entered by the user, or transmits the PAN (Primary Account Number) read from the credit card to the credit card server 200 via the POS device. The POS device works with the credit payment terminal to transmit information such as the payment amount to the credit card server 200. An acquisitioner server may be interposed between the credit processing terminal 70 and the credit card server 200, but for the sake of simplicity, the description of the acquisitioner server will be omitted below. The payment card 60 is, for example, similar in form to a commonly used credit card, with a communication chip embedded in the card base material. The communication chip contains a storage medium that stores the PIN and communicates with an external device via a contactor (or wireless antenna). Alternatively, the payment card 60 may be a magnetic stripe card. Note that the information (messages) transmitted and received when using a credit card includes an authorization message for authentication and a sales message to indicate the payment amount; however, a detailed explanation distinguishing between these will be omitted below.

[0038] The credit card server 200 communicates with the settlement server 100 via a network NW. The credit card server 200 includes, for example, an information management unit 210, a credit interface 220, a settlement distribution unit 230, a credit settlement processing unit 240, and a storage unit 270. Components other than the storage unit 270 are implemented, for example, by a hardware processor such as a CPU executing a program (software). Some or all of these components may be implemented by hardware (including circuitry) such as an LSI, ASIC, FPGA, or GPU, or by the cooperation of software and hardware. The program may be stored in advance in a storage device such as an HDD or flash memory (a storage device with a non-transient storage medium), or it may be stored in a removable storage medium such as a DVD or CD-ROM (a non-transient storage medium) and installed in the storage device when the storage medium is mounted in a drive device. The storage unit 270 stores information such as card user information 272.

[0039] The information management unit 210 edits, adds, and deletes card user information 272 and manages it. Card user information 272 is information that associates, for example, information unique to the user (e.g., PAN), the card payment method, and the user's account ID (used by the payment server 100) with each other. The card payment method is setting information that indicates whether the user will make an electronic payment using their charged balance (balance payment) or a deferred payment in card payments.

[0040] The credit interface 220 determines whether the BIN (Bank Identification Number) in the PAN included in the message received from the credit processing terminal 70 is a code for the company. If it is a code for the company, it passes the message received from the credit processing terminal 70 to the settlement distribution unit 230. If it is not a code for the company, it discards the received message.

[0041] The settlement distribution unit 230 refers to the user's card user information 272 corresponding to the message obtained from the credit interface 220 and determines whether the "card payment method" is set to "post-payment". If the "card payment method" is set to "post-payment", the settlement distribution unit 230 notifies the credit interface 220 of this and passes the message obtained from the credit interface 220 to the credit payment processing unit 240. On the other hand, if the "card payment method" is set to "balance payment", the settlement distribution unit 230 adds the user's account ID to the message obtained from the credit interface 220 and sends it to the settlement server 100 to request electronic payment. The settlement server 100, upon receiving the request for electronic payment, performs the same processing as in S7-1 in Figure 5 and S17-1 in Figure 6.

[0042] The credit interface 220 checks the PAN and expiration date, and verifies whether the cumulative payment amount exceeds the monthly limit. The credit payment processing unit 240 adds the payment amount to the user's monthly usage amount based on the information contained in the message obtained from the payment distribution unit 230, and performs electronic payment by deducting the monthly usage amount from the user's bank account after the closing date.

[0043] [Select compatible channel] The following describes the processing performed by the information processing device. Figure 8 is a diagram illustrating the overview of the processing performed by the information processing device. The first processing unit 140 acquires requests made by the user via voice, text input, etc., through the payment application 20. Then, it performs a process to convert the voice input to text and acquires the resulting text information. The process of converting voice to text may be performed by the payment application 20, by any functional unit of the payment server 100, or by the first processing unit 140.

[0044] The first processing unit 140 inputs input information, including text information, into the first model 194. The first model 194 is a large-scale language model (LLM) built on collective intelligence, which is a knowledge database 193 and crawled data, and is sometimes referred to as a generative AI. The large-scale language model learns from a large amount of text data to understand patterns in human language and performs natural language generation (NLG) tasks. For example, the first model 194 divides the input information into tokens (morphemes), vectorizes them, and outputs first model output information (normalized / abstracted to the extent that the second model 195 can process) that represents the request, based on the contextual understanding obtained by inputting the vectors into a trained model. Furthermore, if the first model 194 determines on its own that the request cannot be sufficiently normalized / abstracted, it may ask the user again via the payment application 20 to concretize the request, and repeat this several times to finalize the first model output information.

[0045] The knowledge database 193 collects, for example, the behavioral history of each user (payment history, charge history, chat history, history of requests made to the electronic payment service (which may include inquiry history), etc.) and attribute information such as age and gender, collected by the payment server 100. The user-specific data contained in the knowledge database 193 is included in the input information, and the statistical data obtained from the knowledge database 193 is used to construct the first model 194. Since the knowledge database 193 contains the personal information of users, the first model 194 is owned independently by the information processing device (including those connected via dedicated lines that are not public lines), unlike large-scale language models that are publicly available on the network. This allows for the reflection of user-specific information in the processing results while protecting the personal information of users.

[0046] Furthermore, the knowledge database 193 may also include estimated user trend information. User trend information is estimated by the estimation unit 150. Figure 9 is a diagram showing an overview of the processing of the estimation unit 150. The estimation unit 150 obtains estimated user trend information by inputting estimation source information, such as the user's behavioral history, into the estimation model. As shown in the figure, the estimation source information includes attribute information such as the user's age and occupation, payment information such as payment time, day of the week, and payment amount, merchant information such as the type of merchant used and payment location, balance information such as deposit and withdrawal history and various balances, P2P history information such as P2P deposit history and withdrawal history (amount / date and time), bank information such as whether or not an account is held, details of loan usage, and whether or not there are any delinquencies, asset management information such as whether or not an account is held, investment experience, and motivation for use, coupon-related information including coupon acquisition and usage history, and electronic flyer information such as the type of merchant related to the viewed electronic flyer and payments made after the electronic flyer was viewed. Trend information includes the user's behavioral age, etc. The behavioral age may be output not as a single value, but as a probability value for each age group. The estimation model is, for example, a model generated by supervised learning using various user information for which trend information is known as training data, and the source information for estimation as learning data. By including the user's trend information in the knowledge database 193, the accuracy of the corresponding channel selection by the second processing unit 160, described later, can be improved. For example, if it is estimated that the user is an elderly person unfamiliar with the operation of the payment app 20, processing such as increasing the probability of human assistance can be performed.

[0047] Furthermore, the knowledge database 193 may include information categorizing the frequency of use of electronic payment services, information classifying user types, and information on estimated user hobbies and preferences. This information may also be estimated by the estimation unit 150.

[0048] The second processing unit 160 inputs at least the output information of the first model into the second model 195, which is constructed as a different model from the first model 194. The second processing unit 160 obtains the second model output information output by the second model 195, which indicates which of the multiple corresponding channels matches the user's request, and selects a corresponding channel that matches the user's request based on the second model output information. The second model 195 may also be input with some information from the knowledge database 193. The second model 195 is, for example, a model constructed by supervised learning using the output information of the first model and some information from the knowledge database 193 as training data, and user feedback (satisfaction with the response) obtained later through surveys, etc., as training data. The second model output information is, for example, vector data with probability values ​​indicating the degree of suitability assigned to each corresponding channel. The second processing unit 160 selects, for example, the corresponding channel with the highest probability value and outputs channel selection information indicating the selected corresponding channel to the response processing unit 170. Furthermore, the second processing unit 160 may pre-acquire the degree of congestion (busyness) for each corresponding channel, and if the corresponding channel with the highest probability value is also congested, it may select the corresponding channel with the second highest probability value.

[0049] The response processing unit 170 activates at least one of several response channels according to the channel selection information. The multiple response channels include, for example, displaying frequently asked questions, a chatbot, automated procedures, coupon issuance, human support, guidance to other service mini-apps, and guidance to other services (at least two of these). For example, if the response processing unit 170 activates the display of frequently asked questions, it transitions the payment app 20 screen to the relevant page; if it activates a chatbot, it activates a chatbot engine (not shown, which may differ from the first model 194) to start a chat with the payment app 20; if it activates automated procedures, it performs automated procedures such as changing the charge limit; if it activates coupon issuance, it activates the coupon issuer's mini-app to display the screen for coupon issuance; and if it activates human support, it arranges for an operator. In addition, if the response processing unit 170 activates guidance to other service mini-apps, it displays, for example, mini-apps provided by other services that link with the electronic payment service (shopping, auctions, flea markets, finance, insurance, securities, real estate, etc.) on the payment app 20. A mini-app is displayed within the screen of the payment app 20, or as a pop-up, and its display content is controlled or determined by the other services mentioned above. By guiding users to other service mini-apps, service collaboration based on the concept of cross-selling can be realized. Furthermore, when the corresponding processing unit 170 initiates guidance to other services, it displays on the payment app 20, for example, a phone number for contacting the other service (same as above) that collaborates with the electronic payment service, or a URL for accessing the other service's website. The first model 194 (first processing unit 140) displays a screen or message on the payment app 20 in accordance with the selection of a corresponding channel by the second processing unit 160.

[0050] By performing the aforementioned processing, the information processing device can interpret the user's request and select the appropriate response channel. Specifically, by configuring the first model 194 and the second model 195 separately, each can be specialized for the required functions, thereby preventing AI hallucination. Furthermore, in constructing the second model 195 for response channel selection, prerequisite information specific to the service to which the information processing device is applied is more necessary than broad information including collective intelligence. By separating this from the construction of the first model 194, it becomes possible to appropriately select the response channel.

[0051] Figures 10 and 11 illustrate the operation of the information processing device. In the example in Figure 10, (1) the user sends a message to the information processing device via the payment app 20 saying, "I can't top up, why?" (2) The first process 140 refers to the user's information in the knowledge database 193 and replies with the message, "It appears that the amount you tried to top up exceeds the daily top-up limit (set by the user). Would you like to increase the limit?" (3) The user sends a message to the information processing device saying, "I see, then please increase the limit by 10,000 yen." (4) The first processing unit 140 outputs first model output information to the second processing unit 160, for example, "Top-up_DAILY_USER_DEFINED_LIMIT error occurred, top-up_daily_user_defined_limit exceeded." (5) The second processing unit 160 selects "automatic procedure" by inputting the first model output information into the second model 195. The corresponding processing unit 170 then automatically increases the user's top-up limit by 10,000 yen. (6) The second processing unit 160 feeds back the result of selecting the corresponding channel to the first processing unit 140. (7) The first processing unit 140 replies with the message, "Understood! The charge limit has been changed from 30,000 yen to 40,000 yen."

[0052] In the example in Figure 11, (1) The user sends a message to the information processing device via the payment app 20 saying, "My card payment is likely to be delayed..." (2) The first processing unit 140 refers to the response rules for the payment server 100 in the knowledge database 193 and replies with the message, "We cannot answer questions regarding card payments. Please contact your card provider. Number: xxx-xxx-xxx" (3) The user sends a message to the information processing device saying, "I don't want to call, is there an email option?" (4) The first processing unit 140 outputs first model output information to the second processing unit 160, for example, "Card inquiries cannot be resolved by the payment server." (5) The second processing unit 160 selects "direct to other services" by inputting the first model output information into the second model 195. The response processing unit 170 then notifies each unit of the telephone number or URL for directing to other services. (6) The second processing unit 160 feeds back the result of the selection of the response channel to the first processing unit 140. (7) The first processing unit 140 replies with the message, "We will introduce the phone number and URL of another service. Phone number: XXX XXX XXX, URL: XXXXXX."

[0053] According to the embodiments described above, it is possible to interpret the user's requests and select an appropriate response channel.

[0054] Although embodiments for carrying out the present invention have been described above using examples, the present invention is not limited in any way to these embodiments, and various modifications and substitutions can be made without departing from the spirit of the present invention. [Explanation of symbols]

[0055] E. Store facilities M medium S Payment System 10. User terminal device 20 Payment Apps 100 Payment Servers 110 Content Provision Department 120 Information Management Department 130 Payment Processing Unit 140 First Processing Unit 150 Estimation section 160 Second Processing Unit 170 Corresponding Processing Unit 190 Memory section 193 Knowledge Database 194 First Model 195 2nd Model

Claims

1. A first processing unit acquires text information that is a text representation of a request issued by a user, and inputs input information including the text information into the first model to acquire first model output information that represents the request output by the first model. A second processing unit inputs at least the output information of the first model to a second model constructed as a different model from the first model, thereby obtaining second model output information indicating which of the multiple corresponding channels output by the second model matches the user's request, and selects a corresponding channel corresponding to the user's request based on the second model output information. An information processing device equipped with the following features.

2. The aforementioned support channels include at least two of the following: providing frequently asked questions, chatbots, automated procedures, coupon distribution, human support, directing users to other service mini-apps, and directing users to other services. The information processing apparatus according to claim 1.

3. The aforementioned input information includes the user's personal information, The first model mentioned above is a large language model (LLM) that is not publicly available on the network. The information processing apparatus according to claim 1.

4. Applicable to electronic payment services, the personal information includes the user's payment history information. The information processing apparatus according to claim 3.

5. Applicable to network services, the personal information includes information on the history of requests made by the user to the network services. The information processing apparatus according to claim 3.

6. It further includes an estimation unit that estimates user trend information, The aforementioned personal information includes estimated trend information of the aforementioned user, The information processing apparatus according to claim 3.

7. The aforementioned trend information includes behavioral age, The information processing apparatus according to claim 6.

8. The first model further outputs the first model output information based on collective intelligence obtained on the network. The information processing apparatus according to claim 1.

9. Before outputting the output information of the first model, the first model queries the user to obtain the text information again. The information processing apparatus according to claim 1.

10. The second processing unit inputs, in addition to the output information of the first model, at least a portion of the information input to the first model to the second model. The information processing apparatus according to claim 1.

11. Information processing device, A first process involves obtaining text information that represents the request issued by the user, inputting the input information including the text information into the first model, and thereby obtaining first model output information that represents the request output by the first model. A second process involves inputting at least the output information of the first model into a second model constructed as a different model from the first model, thereby obtaining second model output information that indicates which of the multiple corresponding channels output by the second model matches the user's request, and selecting a corresponding channel that corresponds to the user's request based on the second model output information. An information processing method that performs the following.

12. In an information processing device, A first process involves obtaining text information that represents the request issued by the user, inputting the input information including the text information into the first model, and thereby obtaining first model output information that represents the request output by the first model. A second process involves inputting at least the output information of the first model into a second model constructed as a different model from the first model, thereby obtaining second model output information that indicates which of the multiple corresponding channels output by the second model matches the user's request, and selecting a corresponding channel that corresponds to the user's request based on the second model output information. A program to execute.

Images