Information processing systems, information processing methods, and programs

Abstract

The system generates a draft reply based on the review content, control conditions, and past cases, and the finalized reply content can be used to generate subsequent reply drafts. It can also be applied to generating response drafts and subsequent processing for reservation requests, inquiries, and other target input information. [Solution] The information processing system acquires the review content corresponding to the target review, acquires one or more control conditions related to the target review, identifies reply policy information according to the control conditions, and acquires case information, including past review content related to the review content and corresponding reply content, from a case storage unit associated with a store, business, or account. Furthermore, the information processing system generates a draft reply based on the review content, reply policy information, and case information, acquires the confirmed reply content based on the draft reply, and stores the confirmed reply content as case information that can be used to generate subsequent draft reply texts.

Description

Technical Field

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

Background Art

[0002] In recent years, reviews or word-of-mouth comments on products or services have been continuously posted, and store owners, business operators, or operators of various accounts are required to manage these reviews or word-of-mouth comments and respond as needed. For this reason, various techniques for listing, selecting, or displaying reviews or word-of-mouth comments in an easy-to-view manner have been proposed.

[0003] For example, Patent Document 1 discloses a word-of-mouth management device that accepts from a user a selection of a representative word-of-mouth comment representing a plurality of word-of-mouth comments and highlights the selected representative word-of-mouth comment. Patent Document 1 controls the highlighting target according to the number of selected word-of-mouth comments and selects a representative word-of-mouth comment to be highlighted based on the posting date and time, event date, or height of evaluation, etc., making it easier for service providers to manage the word-of-mouth comments they want to show to customers.

[0004] Also, Patent Document 1 shows a list of search results of word-of-mouth comments, a word-of-mouth comment response screen, a poster confirmation screen, a reply comment input screen, and a representative word-of-mouth comment setting screen, and it is also described that the input of a reply comment may be included as a response to the word-of-mouth comment. However, the main focus of Patent Document 1 is on the selection, highlighting, and display management of word-of-mouth comments, and it is not configured as a technique for generating the reply text itself.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, in actual review management, it is crucial not only to manage which reviews or testimonials are displayed prominently, but also to quickly draft appropriate responses for each individual review. In particular, if the tone, expression style, considerations, or operational policies of responses differ from store to store, business to business, or account to account, simply making reviews or testimonials viewable is not sufficient in practice.

[0007] In this regard, while Patent Document 1 shows a screen in which reply comments to reviews can be entered, it does not disclose a configuration that automatically or semi-automatically generates a reply draft according to the target review by acquiring control conditions such as industry, evaluation expression, and platform type in addition to the review content, and identifying reply policy information according to these conditions.

[0008] Furthermore, Patent Document 1 does not disclose a configuration that acquires past review content and corresponding reply content associated with stores, businesses, or accounts as case information, and generates reply drafts using said case information. Therefore, it is difficult to generate reply drafts for new reviews while maintaining consistency with actual replies given in the past.

[0009] Furthermore, in the practical application of review replies, users may either revise the generated reply draft before finalizing it, or they may finalize the generated reply draft as is. However, conventional technology did not adequately consider the systematic storage and reuse of such finalized reply content as case information that can be used to generate subsequent reply drafts. Consequently, it was not possible to implement continuous improvements that would make it easier to generate reply drafts suitable for the store, business, or account as the system was used repeatedly.

[0010] Thus, while conventional technology is useful for managing the display of reviews or word-of-mouth, there is still room for improvement in terms of generating response drafts for target reviews, utilizing past examples, and reusing finalized response content. Therefore, the present invention aims to generate response drafts suitable for target reviews in accordance with review response practices, and to enable the use of finalized response content for future response draft generation. [Means for solving the problem]

[0011] According to one aspect of the present invention, an information processing system is provided. The information processing system has one or more control units, each of which acquires review content corresponding to a target review, acquires one or more control conditions related to the target review, identifies reply policy information according to the control conditions, acquires case information including past review content related to the review content and corresponding reply content from a case storage unit associated with a store, business operator, or account, generates a draft reply based on the review content, reply policy information and case information, acquires the confirmed reply content based on the draft reply, and stores the confirmed reply content as case information that can be used to generate subsequent draft reply texts. [Brief explanation of the drawing]

[0012] [Figure 1] Figure 1 shows an example of the system configuration of an information processing system. [Figure 2] Figure 2 shows an example of the hardware configuration of a user terminal. [Figure 3] Figure 3 shows an example of the hardware configuration of the management server. [Figure 4] Figure 4 shows an example of the functional configuration of a user terminal. [Figure 5] Figure 5 shows an example of the functional configuration of the management server. [Figure 6] Figure 6 shows an example of a flowchart for information processing related to the generation of reply drafts and case registration. [Figure 7]Figure 7 shows an example of an information processing flowchart that retrieves review content based on the specified screen area, identifies the reply input area, and provides a draft reply for input. [Modes for carrying out the invention]

[0013] The embodiments will be described below with reference to the drawings. The various features shown in the embodiments below can be combined with each other.

[0014] <Embodiment 1> 1. System Configuration Figure 1 shows an example of the system configuration of the information processing system 1000. As shown in Figure 1, the information processing system 1000 may include a user terminal 20 and a management server 30. Furthermore, the information processing system 1000 may include an external generation and processing device 40 as needed.

[0015] The user terminal 20, the management server 30, and the external generation processing device 40 may be connected to each other via a network 50. The network 50 may be a wired or wireless communication network and may include, for example, a wide area network (WAN), a local area network (LAN), the internet, an enterprise network, a mobile communication network, or any combination thereof.

[0016] The user terminal 20 may be an information processing device used by a store, business operator, account manager, person in charge, or other user. The user terminal 20 may be, for example, a personal computer (PC), notebook PC, tablet terminal, smartphone, etc. In this embodiment, the user terminal 20 may have an extension module, add-on, plugin, or dedicated application installed that provides support for responding to reviews or word-of-mouth. The user terminal 20 may operate on a review management interface displayed on a web browser and perform main processes such as acquiring review content, acquiring control conditions, identifying response policy information, displaying draft response text, acquiring confirmed response content, and providing input support to the response input area.

[0017] The review management interface may be, for example, a screen for an online review service, a store information service, a reservation service, an inquiry management service, or a customer contact management service. The review management interface may display an area for displaying the target review, a reply input area, evaluation information, posting information, or platform type information. The user terminal 20 may obtain the review content, evaluation expression, platform type, and other information corresponding to the target review from the screen displayed on the web browser. In this embodiment, reply support for reviews or word-of-mouth is described as an example, but the processing target is not limited to this, and it may also be applied to response support or subsequent processing support for user input information such as reservation requests, inquiries, free-response questionnaires, message bodies, etc.

[0018] The management server 30 may be a device that stores various data used in the information processing system 1000 and provides those data in response to requests from the user terminal 20. The management server 30 may store, for example, user authentication information, contract plan information, industry type information, language setting information, information regarding the target language for learning, case information, and other setting information. Further, the management server 30 may have a case storage unit that associates, for each store, business operator, or account, past review content with the reply content corresponding to the past review content.

[0019] The user terminal 20 may obtain review content corresponding to the target review, obtain one or more control conditions related to the target review, and identify reply policy information according to the control conditions while referring to the information stored in the management server 30. The control conditions may include, for example, industry type information, evaluation expression information, and platform type information, and may further include review language information, user language information, regional information, legal risk information, policy compliance information, and other additional condition information.

[0020] The user terminal 20 may obtain case information related to the review content from the case information stored in the management server 30 and generate a reply text based on the review content, reply policy information, and case information. The generation of the reply text may be executed by the user terminal 20, or may be realized by the user terminal 20 sending a generation request to an external generation processing device 40 and receiving the generation result.

[0021] The external generation processing device 40 may be an external service that provides a generation model, language model, or inference application programming interface (API) for generating reply drafts. The user terminal 20 may provide the external generation processing device 40 with input information based on review content, control conditions, reply policy information, and case information, and obtain a reply draft. The management server 30 is mainly responsible for storing case information, configuration information, authentication information, and usage status information, and the main control for generating reply drafts may be performed on the user terminal 20 side.

[0022] The user terminal 20 may display the generated reply draft and accept operations from the user for confirmation, editing, or confirmation. The user terminal 20 may then retrieve the confirmed reply content based on the reply draft and send the confirmed reply content to the management server 30. The management server 30 may store the received confirmed reply content as example information that can be used to generate subsequent reply drafts.

[0023] In one embodiment, the user terminal 20 may send the confirmed reply content to a platform providing a review management interface or an external system that cooperates with it. The sending of the reply content may be performed in response to a sending operation by the user, or it may be performed automatically according to the platform specifications, contract terms, permission settings, or operational rules. For example, the user terminal 20 may identify the reply content reflected in the reply input area as the target for sending, and output a sending request to the platform when predetermined execution conditions are met.

[0024] In one embodiment, the user terminal 20 may identify the confirmed reply content as the recipient. The user terminal 20 may automatically execute a request to send the confirmed reply content to the platform or an external system cooperating with it when predetermined execution conditions are met.

[0025] On the other hand, if automatic transmission is not permitted due to platform specifications, contract terms, or operational rules, the user terminal 20 may only reflect the reply content in the reply input area or present it as an input option, leaving the transmission operation itself to the user. This allows the information processing system 1000 to support both modes of operation, including those with automatic transmission and those without, depending on the constraints of each platform.

[0026] In one embodiment, the user terminal 20 may calculate the difference between the generated reply draft and the finalized reply content based on the reply draft. The difference may be calculated based on the difference in character count, editing distance, amount of added text, amount of deleted text, amount of replacement, or other indicators. The user terminal 20 may determine whether the difference amount satisfies predetermined conditions, and if the predetermined conditions are met, it may store the finalized reply content as example information in the management server 30.

[0027] In one embodiment, the user terminal 20 may generate or display a draft reply in a first language corresponding to the language of the review content, and a draft display text in a second language corresponding to the language set on the user terminal 20. Alternatively, the management server 30 may store only the confirmed reply content corresponding to the language set as the target language for learning as example information.

[0028] In one embodiment, if multiple reviews are managed according to a predetermined processing order, the user terminal 20 may sequentially generate a draft reply for each review according to the processing order. This enables continuous reply support for multiple reviews.

[0029] In one embodiment, the user terminal 20 may accept input to obtain review content from within a screen area specified by the user. Alternatively, the user terminal 20 may identify an editable element adjacent to the screen area as a reply input area and provide the user with a draft reply to input into the reply input area. This allows for reply support not only for standard review management interfaces but also for reviews or similar information displayed on any screen.

[0030] The main processes described below are assumed to be performed by the user terminal 20, but are not limited to this. For example, at least part of the process of obtaining review content, obtaining control conditions, identifying reply policy information, obtaining case information, generating reply drafts, obtaining confirmed reply content, and assisting with input into the reply input area may be assisted by the external generation processing device 40. In addition, the management server 30 may be configured as a device that primarily performs storage processing, authentication processing, and data provision processing.

[0031] The information processing system described in the claims may consist of multiple devices or a single device. For example, some or all of the functions of the user terminal 20 and the management server 30 may be implemented within the same device, or the functions of the external generation processing device 40 may be incorporated into the management server 30 or the user terminal 20.

[0032] 2. Hardware Configuration (1) Hardware configuration of user terminal 20 Figure 2 shows an example of the hardware configuration of a user terminal 20. As shown in Figure 2, the user terminal 20 includes, as a hardware configuration, a control unit 210, a storage unit 220, an input unit 230, an output unit 240, a communication unit 250, and an internal bus 260. The control unit 210, the storage unit 220, the input unit 230, the output unit 240, and the communication unit 250 are electrically connected via the internal bus 260.

[0033] The control unit 210 is one of the following: a Central Processing Unit (CPU), a Micro Processing Unit (MPU), a System-on-a-Chip (SoC), a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), a Graphics Processing Unit (GPU), or a combination of at least two of these, and controls the entire user terminal 20 and executes processing in response to information input from, for example, the input unit 230, the communication unit 250, or the storage unit 220.

[0034] The storage unit 220 is one of the following: Read Only Memory (ROM), Random Access Memory (RAM), Flash Memory, Solid State Drive (SSD), or a combination of at least two of these, and stores the program and data used by the control unit 210 when it executes processing based on the program. The storage unit 220 may store programs corresponding to extension modules, add-ons, plugins, or dedicated applications that run on a web browser, review content, evaluation information, platform type information, reply policy information, generated reply drafts, confirmed reply content, difference amount, screen area specification information, reply input area specification information, and other temporary processing data. The storage unit 220 is an example of a non-temporary recording medium that can be read by the computer on which the program is recorded.

[0035] In this specification, the data used by the control unit 210 when executing processing based on the program is described as being stored in the storage unit 220, but it may also be stored in the storage unit of another device that can communicate with the user terminal 20. The data may be stored in the storage unit of any device as long as the control unit 210 can access and / or retrieve it.

[0036] The control unit 210 executes processing based on the program stored in the memory unit 220, thereby realizing the processing shown in the functional configuration and flowcharts described later. For example, the control unit 210 can perform at least some of the following: obtaining review content from the review management interface, obtaining control conditions, identifying reply policy information, referring to case information obtained from the management server 30, generating or requesting the generation of a reply draft, displaying the reply draft, obtaining confirmed reply content, calculating the difference amount, identifying the reply input area, and providing reply input support. In this embodiment, the main processing for reply support is described as being performed on the user terminal 20, but is not limited to this.

[0037] The input unit 230 is a device that inputs information to the user terminal 20 in response to user operations. The input unit 230 receives operation inputs made by the user. The operation inputs are transferred to the control unit 210 via the internal bus 260 as command signals. The control unit 210 can, if necessary, perform predetermined controls and / or calculations based on the transferred command signals. The input unit 230 may be included in the casing of the user terminal 20 or it may be externally mounted. For example, the input unit 230 may be implemented as a touch panel integrated with the output unit 240. When the input unit 230 is implemented as a touch panel, the user can input tap operations, swipe operations, drag operations, etc. to the input unit 230. The input unit 230 may be a keyboard, mouse, trackpad, switch buttons, etc. instead of a touch panel.

[0038] The output unit 240 is a display unit, such as a display, that outputs information as a screen of a graphical user interface (GUI) that can be operated by the user. The output unit 240 may be included in the housing of the user terminal 20 or it may be an external device. More specifically, the output unit 240 may be implemented as a display device such as a liquid crystal display, an organic electroluminescent display, or a plasma display. The output unit 240 may display various information such as a review management interface, reply drafts, display drafts, progress information, candidate example information, difference confirmation information, and more.

[0039] The communication unit 250 is a communication interface that connects the user terminal 20 to the network 50 and mediates communication with other devices. The communication unit 250 can be implemented, for example, by hardware that realizes wired communication and / or wireless communication, and circuits that control them. Based on instructions from the control unit 210, the communication unit 250 performs data transmission and reception processing via the network 50 according to predetermined communication protocols such as Transmission Control Protocol (TCP) / Internet Protocol (IP), User Datagram Protocol (UDP), Hypertext Transfer Protocol (HTTP), and HTTP over Transport Layer Security (HTTPS). For example, the communication unit 250 can send and receive authentication information, contract plan information, industry information, language setting information, case information, confirmed reply content, usage status information, etc. with the management server 30. In addition, depending on the embodiment, the communication unit 250 can send and receive generation requests and reply drafts with the external generation processing device 40.

[0040] The hardware configuration of the user terminal 20 is not limited to the example described above. For example, the user terminal 20 may include multiple control units 210, or the storage unit 220 and communication unit 250 may be implemented in a distributed manner by multiple devices. Furthermore, the user terminal 20 is not limited to a personal computer, but may be a notebook computer, tablet terminal, smartphone, or other portable or stationary information terminal.

[0041] (2) Hardware configuration of the management server 30 Figure 3 shows an example of the hardware configuration of the management server 30. As shown in Figure 3, the management server 30 includes a control unit 310, a storage unit 320, a communication unit 330, and an internal bus 340 as its hardware configuration. The control unit 310, the storage unit 320, and the communication unit 330 are electrically connected via the internal bus 340.

[0042] The control unit 310 is one of the following: a central processing unit, a microprocessing unit, a system-on-a-chip, a field-programmable gate array, an application-specific integrated circuit, a graphics processing unit, or a combination of at least two of these, and controls the entire management server 30 and executes processing in response to information input from, for example, a user terminal 20 or a storage unit 320.

[0043] The storage unit 320 is one of the following: a hard disk drive (HDD), read-only memory, random access memory, solid-state drive, or a combination of at least two of these, and stores data used by the program and control unit 310 when executing processing based on the program. The storage unit 320 may store user authentication information, contract plan information, setting information associated with stores, businesses or accounts, industry information, language setting information, information on target languages ​​for learning, usage information, and case information associated with past review content and corresponding replies. The storage unit 320 may also function as a case storage unit. Various types of data may be managed in relational databases, key-value stores, vector databases, or any other format.

[0044] The management server 30 may be configured as a device primarily responsible for memory processing, authentication processing, and data provision processing. For example, the control unit 310 may read and provide case information, setting information, or authentication-related information in response to a request from the user terminal 20, receive the confirmed reply content sent from the user terminal 20 and store it in the storage unit 320, and update the usage status information as necessary. On the other hand, the main processing such as acquiring review content, acquiring control conditions, identifying reply policy information, generating reply drafts, calculating difference amounts, and providing reply input support may be performed by the user terminal 20.

[0045] The communication unit 330 is a communication interface that connects the management server 30 to the network 50 and mediates communication with other devices. Based on instructions from the control unit 310, the communication unit 330 performs data transmission and reception processing via the network 50 according to a predetermined communication protocol. For example, the communication unit 330 can send and receive authentication information, contract plan information, industry information, language setting information, case information, confirmed reply content, and usage status information with the user terminal 20.

[0046] The hardware configuration of the management server 30 is not limited to the example above. For example, the management server 30 may include multiple control units 310, and the storage unit 320 may be implemented as a single storage device or distributed across multiple storage devices. Similarly, the communication unit 330 may have multiple communication interfaces.

[0047] In this specification, a configuration is illustrated in which the user terminal 20 performs the main processing and the management server 30 primarily performs storage processing, authentication processing, and data provision processing. However, the entity executing the processing is not limited to this configuration. For example, the management server 30 may perform some of the processing that the user terminal 20 performs, or the user terminal 20 may perform some of the processing that the management server 30 performs.

[0048] 3. Functional Configuration Figure 4 shows an example of the functional configuration of the user terminal 20. As shown in Figure 4, the user terminal 20 may include a local storage unit 410, a review acquisition unit 420, a control condition acquisition unit 430, a reply policy identification unit 440, a case information acquisition unit 450, a reply draft generation unit 460, a confirmed reply acquisition unit 470, a case registration control unit 480, and a display control unit 490. Furthermore, in one embodiment, the user terminal 20 may also include a screen area analysis unit 495.

[0049] Each of these functional units is a functional block realized by, for example, the control unit 210 executing a program stored in the storage unit 220 and controlling hardware such as the input unit 230, output unit 240, and communication unit 250. In this embodiment, the main processes such as acquiring review content, acquiring control conditions, identifying reply policy information, generating a reply draft, acquiring confirmed reply content, and reply input support will be described as being performed by the user terminal 20.

[0050] The local memory unit 410 may store review content, control conditions, reply policy information, generated reply draft, finalized reply content, difference amount, display draft, and other information during processing. The local memory unit 410 may also store cache, setting information, operation history, or screen analysis results that are temporarily stored on the user terminal 20.

[0051] In this specification, "local storage unit 410" refers to the logical storage means in the functional configuration shown in Figure 4. The local storage unit 410 can be implemented in the storage unit 220 with the hardware configuration shown in Figure 2 or in an external storage device that can communicate with it. That is, the local storage unit 410 may be implemented on the same physical resources as the storage unit 220 as hardware, or it may be implemented distributed across multiple physical storage resources.

[0052] In this specification, "review content" refers to information that may include, at least in part, strings of text displayed as reviews or testimonials, evaluation expressions, posting date and time, poster attributes, display information on the platform, and other relevant information. Furthermore, "control conditions" refers to conditional information used to identify reply policy information or generate reply drafts. Control conditions may include, for example, industry information, evaluation expression information, platform type information, review language information, user language information, regional information, legal risk information, policy compliance information, and other additional conditional information.

[0053] In this specification, "Target Input Information" refers to information provided by the user that is subject to the generation of response drafts, guidance drafts, confirmation drafts, or subsequent processing proposals. Target Input Information is not limited to reviews or word-of-mouth, but may include reservation requests, inquiries, free-response questionnaires, chat messages, email bodies, or similar input information. Review content is an example of Target Input Information.

[0054] In this specification, "response draft" refers to a draft text that shows the content of the response generated in response to the target input information. A response draft is not limited to a response to a review or word of mouth, but may include a notification text for a reservation request, a response text for an inquiry, a confirmation request text, or a processing text or other response text that is passed to a business system.

[0055] In this specification, "target review" refers to a review or comment selected as the target for generating a response. The target review may be a review selected by the user, a review currently being displayed, or a review selected from multiple reviews according to a predetermined processing order.

[0056] In this specification, “evaluation representation information” means information indicating an evaluation assigned to or associated with the review in question. Evaluation representation information may include, for example, star ratings, icon ratings, numerical ratings, graded ratings, and other forms of evaluation display.

[0057] In this specification, "platform type information" refers to information indicating the type of service, medium, or screen from which the target review is obtained. Platform type information may include, for example, the type of review service, the type of store information service, the type of reservation service, the type of inquiry management service, or the type of customer contact management service.

[0058] In this specification, "additional axis" refers to additional conditional axes used to identify response policy information or generate response drafts, in addition to industry information, evaluation expression information, and platform type information. "Additional conditional information" refers to conditional information corresponding to the additional axis. Additional conditional information may include, for example, review language information, user language information, regional information, legal risk information, policy compliance information, and other conditional information.

[0059] In this specification, "display text" refers to the text displayed to the user to help them understand the content of the reply text. The display text may be a translation of the reply text, a summary, an explanatory text, or a confirmation text.

[0060] In this specification, "language to be learned" refers to the language in which confirmed reply content is stored as example information and used for generating subsequent reply drafts. The language to be learned may be determined based on user settings, account settings, or operational rules.

[0061] In this specification, "proximity" may be a concept indicating the degree to which the elements in question are related on screen, in terms of document structure, or functionally. Proximity is not limited to absolute physical distance, but may also be determined based on relative positional relationships, whether or not they are within a predetermined distance, the amount of overlap, vertical and horizontal positioning relationships, hierarchical relationships in the document object model, element attributes, label strings, placeholder strings, correspondence with buttons or headings, or at least two combinations thereof.

[0062] In one embodiment, proximity may be represented by proximity, fit, or relevance calculated for multiple candidate elements. Proximity, fit, or relevance may be calculated based, for example, on the vertical distance to the screen area, the amount of horizontal overlap, whether or not it is located below the screen area, its display state, its editable state, its hierarchical relationship on the document object model, its positional relationship to label text, placeholder text, buttons, or headings.

[0063] In one embodiment, proximity may be determined by being below a predetermined distance, by selecting the element that is relatively closest among multiple candidate elements, or by using a combination of the predetermined distance condition and relative comparison. Therefore, "proximity" is not limited to being within a certain distance, but may also include cases where the relative relevance is high among multiple candidates.

[0064] Furthermore, proximity may be determined based on a single indicator, or it may be determined comprehensively by weighting multiple indicators. This allows for flexible identification of the reply input area corresponding to the review content, even when the screen structure, platform structure, or display mode differs.

[0065] The review acquisition unit 420 acquires the review content corresponding to the target review displayed on the review management interface. For example, the review acquisition unit 420 may acquire the body text, evaluation expression, posting information, and other information of the target review from the review management interface displayed on a web browser. In one embodiment, the review acquisition unit 420 may acquire the review content corresponding to the review selected by the user, the currently displayed review, or the review selected according to the processing order.

[0066] The control condition acquisition unit 430 acquires one or more control conditions related to the target review. For example, the control condition acquisition unit 430 may acquire control conditions based on the evaluation expression or platform type displayed on the review management interface, and may acquire industry information, user language information, and other condition information based on the setting information stored in the user terminal 20 or the management server 30.

[0067] In this specification, "reply policy information" refers to information that defines the contentual direction of the reply. Reply policy information may include, for example, information indicating a policy centered on gratitude, a policy centered on apology and intention to improve, a policy including encouragement of return visits, a policy prioritizing concise guidance, or a policy prioritizing legal or policy considerations.

[0068] The reply policy identification unit 440 identifies reply policy information according to the acquired control conditions. For example, the reply policy identification unit 440 may identify reply policy information according to a combination of industry information, evaluation expression information, and platform type information. Alternatively, the reply policy identification unit 440 may refer to additional condition information to identify reply policy information that takes into account language, region, risk, or policy compliance.

[0069] The case information acquisition unit 450 acquires case information related to the review content from the case information stored in the management server 30. The case information may include past review content associated with stores, businesses, or accounts, and corresponding reply content. In one embodiment, the case information acquisition unit 450 may acquire past review content similar to the review content and the reply content actually sent to that past review content.

[0070] The reply draft generation unit 460 generates a reply draft based on the review content, reply policy information, and case information. The reply draft generation unit 460 may generate the reply draft within the user terminal 20, or it may send a generation request to an external generation processing device 40 via the communication unit 250 and obtain a reply draft in response. In one embodiment, the reply draft generation unit 460 may generate a reply draft in a first language corresponding to the language of the review content and a display draft in a second language corresponding to the language set on the user terminal 20.

[0071] The confirmed reply acquisition unit 470 acquires the confirmed reply content based on the reply draft. The confirmed reply content may be the reply content that the user has edited the reply draft to, or it may be the reply content that the user has confirmed without editing the reply draft. The confirmed reply acquisition unit 470 may also acquire the content entered in the reply input area, the content that was the subject of the confirmation operation, or the content that was finally selected as the target for transmission.

[0072] In one embodiment, the confirmed reply acquisition unit 470 may identify the confirmed reply content as the recipient of transmission. Alternatively, the display control unit 490 or the confirmed reply acquisition unit 470 may automatically execute a transmission request for the confirmed reply content to the platform or an external system cooperating therewith when predetermined execution conditions are met.

[0073] The case registration control unit 480 controls the registration of the finalized reply content as case information that can be used to generate subsequent reply drafts. For example, the case registration control unit 480 may calculate the difference between the generated reply draft and the finalized reply content, and if the difference amount satisfies predetermined conditions, it may send the case information, including the review content and the finalized reply content, to the management server 30 for storage. The difference amount may be calculated based on at least one of the following: character difference, edit distance, added text amount, deleted text amount, replacement amount, etc.

[0074] The display control unit 490 controls the display of various information, such as reply drafts, display drafts, case information, progress information, etc., on the review management interface or a related screen. The display control unit 490 may display a reply draft corresponding to a single review, or, if multiple reviews are managed according to a predetermined processing order, it may sequentially display reply drafts for each review according to the processing order.

[0075] The screen area analysis unit 495 may analyze the screen area specified by the user, obtain review content from within the screen area, and identify editable elements adjacent to the screen area as reply input areas. For example, the screen area analysis unit 495 may identify the review content and reply input areas using element analysis based on the Document Object Model (DOM), text node extraction, positional relationship analysis, or proximity determination. The display control unit 490 or the confirmed reply acquisition unit 470 may provide input assistance to the reply input areas identified by the screen area analysis unit 495.

[0076] Figure 5 shows an example of the functional configuration of the management server 30. As shown in Figure 5, the management server 30 may include a server storage unit 510, an authentication management unit 520, a configuration information management unit 530, a case information provision unit 540, a case information registration unit 550, and a usage status management unit 560.

[0077] Each of these functional units is a functional block realized, for example, by the control unit 310 executing a program stored in the storage unit 320 and controlling hardware such as the communication unit 330. In this embodiment, the management server 30 is described as primarily responsible for storage processing, authentication processing, and data provision processing.

[0078] The server storage unit 510 stores user authentication information, contract plan information, industry information, language setting information, information on target languages ​​for learning, usage information, and case information. The case information may include past review content and corresponding replies associated with each store, business, or account.

[0079] In this specification, "server storage unit 510" refers to the logical storage means in the functional configuration shown in Figure 5. The server storage unit 510 may be implemented in the storage unit 320 with the hardware configuration shown in Figure 3 or in an external storage device that can communicate with it. The server storage unit 510 may be implemented on a single physical resource or distributed across multiple physical storage resources.

[0080] In this specification, "case memory unit" refers to a storage means for storing case information associated with a store, business operator, or account. The case memory unit may be implemented on the storage unit 320 of the management server 30, or it may be implemented distributed across multiple storage resources, including an external storage device.

[0081] The authentication management unit 520 may perform user authentication based on authentication information transmitted from the user terminal 20. The configuration information management unit 530 may manage industry information, language setting information, contract plan information, information on target languages ​​for learning, and other configuration information provided to the user terminal 20. The case information provision unit 540 may search for or extract and provide case information related to the review content in response to a request from the user terminal 20.

[0082] The case information registration unit 550 may receive the confirmed reply content sent from the user terminal 20 and register it as case information in association with the review content. In one embodiment, the case information registration unit 550 may register only confirmed reply content corresponding to the language set as the target language for learning. The usage status management unit 560 may manage the number of times the reply draft generation function is used, the usage status for each plan, and other operational information.

[0083] The functional units of the user terminal 20 and the management server 30 are not limited to the examples described above. For example, the management server 30 may perform some of the functions that the user terminal 20 performs, or the user terminal 20 may perform some of the functions that the management server 30 performs. Furthermore, the external generation processing device 40 may be illustrated as a separate functional block, or it may be treated as part of the reply text generation unit 460.

[0084] 4. Information Processing The information processing of Embodiment 1 will be described below.

[0085] (1) Overview of the process The user terminal 20 (or the information processing system 1000 including the management server 30 and the external generation processing device 40) acquires the review content corresponding to the target review, acquires one or more control conditions related to the target review, identifies reply policy information according to the control conditions, acquires case information from the case information stored in the management server 30 that includes past review content related to the review content and corresponding reply content, and generates a reply draft based on the review content, reply policy information and case information. Furthermore, the information processing system 1000 acquires the confirmed reply content based on the reply draft and stores the confirmed reply content as case information that can be used to generate subsequent reply drafts.

[0086] In this embodiment, for the sake of explanation, the focus will be on generating response drafts to target reviews. However, the information processing system 1000 is not limited to acquiring reviews or word-of-mouth information. For example, the information processing system 1000 may acquire reservation requests, inquiries, free-form questionnaire responses, message text, and other target input information, generate response drafts based on control conditions, response policy information, and case information, and store the finalized response content as case information that can be used for subsequent draft generation.

[0087] Furthermore, in one embodiment, the information processing system 1000 may perform subsequent actions on a reservation management system, inquiry management system, customer support system, or business support system based on the confirmed response content, or link information necessary for subsequent actions. This makes it possible to provide comprehensive support for user feedback, from draft generation to case learning and subsequent business processing.

[0088] Here, "retrieving review content" means obtaining at least a portion of the text, rating expressions, posting information, language information, and other elements displayed as a review or testimonial. The method of obtaining review content is not limited; it may be obtained from reviews displayed on the review management interface, or from a screen area specified by the user.

[0089] Furthermore, "acquiring control conditions" means acquiring conditional information used to identify reply policy information or generate a draft reply. Control conditions may include, for example, industry information, evaluation expression information, and platform type information. In addition, control conditions may include review language information, user language information, regional information, legal risk information, policy compliance information, and other additional conditional information.

[0090] Furthermore, "identifying reply policy information" means identifying information that defines the contentual direction of the reply draft, depending on the control conditions. Reply policy information may include, for example, information indicating a policy centered on gratitude, a policy centered on apology and intention to improve, a policy including encouraging return visits, a policy prioritizing concise guidance, or a policy prioritizing legal or policy considerations.

[0091] Furthermore, "acquiring case information" means acquiring information including past review content and corresponding replies associated with a store, business, or account. Case information may also include past review content similar to the current review and replies actually sent to that past review content.

[0092] Furthermore, "retrieving the finalized reply" means retrieving the reply that was ultimately adopted after the user reviewed, edited, or finalized the generated reply draft. The finalized reply may be the reply that the user edited before finalizing it, or it may be the reply that was finalized without any editing.

[0093] By performing this process, the information processing system 1000 can generate a draft reply based not only on the review content but also on control conditions and past cases. Furthermore, by storing the finalized reply content as case information, the operational results can be reflected in the generation of subsequent draft replies, regardless of whether or not the user has edited them. This makes it possible to improve the efficiency of reply creation work and the quality of replies while maintaining reply trends for each store, business, or account.

[0094] (2) Details of information processing The details of the information processing in this embodiment will be described below with reference to Figures 6 and 7. Figure 6 is an example of a flowchart of information processing related to the generation of reply drafts and registration of cases. Figure 7 is an example of a flowchart relating to an embodiment in which review content is obtained using a screen area specified by the user, a reply input area is identified, and a reply draft is provided for input.

[0095] In this specification, "storing case information" means associating the review content with the final response content adopted in response to the review content and storing it in a format that can be used for generating response drafts in the future. The case information may be stored in the management server 30, or it may be temporarily stored in the user terminal 20 and then sent to the management server 30.

[0096] In this specification, "difference amount" refers to information indicating the difference between the generated reply draft and the finalized reply content. The difference amount can be calculated based on at least one of the following: character difference, edit distance, amount of added text, amount of deleted text, amount of replaced text, etc.

[0097] In this specification, "prescribed conditions" refer to conditions for determining whether or not to register the finalized reply content as case information based on the difference amount. Prescribed conditions may include, for example, that the difference amount is greater than or equal to a predetermined threshold, that the difference amount is less than or equal to a predetermined threshold, that the difference amount is within a predetermined range, or that the difference amount corresponds to a predetermined editing mode. The threshold may be set based on at least one of the following: the difference in character count, editing distance, amount of added text, amount of deleted text, or amount of replacement. Furthermore, prescribed conditions may include conditions that restrict the scope of case registration based on the setting of enabling or disabling learning set for each store, business operator, or account, review type, evaluation category, frequency of use, period, legal risk information, or policy compliance information. However, prescribed conditions are not limited to these and may include other conditions set based on operational policies, business rules, or estimation results of machine learning models.

[0098] In this specification, "predetermined processing order" refers to the order in which response drafts are generated or responses are processed for multiple reviews. The predetermined processing order may be in order of submission date and time, acquisition date, importance, rating, unaddressed period, or user-specified order, etc.

[0099] (Step S601) The review acquisition unit 420 acquires the review content corresponding to the target review. The method of acquiring the review content is not limited and may be acquired from the review text, evaluation expression, posting information, or other displayed information displayed on the review management interface. The review acquisition unit 420 then proceeds to step S602.

[0100] (Step S602) The control condition acquisition unit 430 acquires one or more control conditions related to the target review. For example, the control condition acquisition unit 430 may acquire industry information, evaluation expression information, and platform type information. In addition, the control condition acquisition unit 430 may acquire review language information, user language information, regional information, legal risk information, policy compliance information, and other additional condition information as needed. The evaluation information may include at least one of the following: a star rating, an icon rating, a numerical rating, or a graded rating. The control condition acquisition unit 430 then proceeds to step S603.

[0101] (Step S603) The reply policy identification unit 440 identifies reply policy information according to the control conditions obtained in step S602. For example, the reply policy identification unit 440 may identify reply policy information according to a combination of industry information, evaluation expression information, and platform type information. Alternatively, it may identify reply policy information that takes into account language, region, risk, or policy compliance by referring to additional condition information. The reply policy identification unit 440 then proceeds to step S604.

[0102] (Step S604) The case information acquisition unit 450 acquires case information related to the review content from the case information stored in the management server 30. For example, the case information acquisition unit 450 may acquire past review content similar to the review content and the actual replies sent to that past review content. The method for determining relevance or similarity is not limited and may be performed by keyword matching, similarity determination based on embedding vectors, matching of category information, or a combination thereof. The case information acquisition unit 450 then proceeds to step S605.

[0103] (Step S605) The reply draft generation unit 460 generates a reply draft based on the review content, reply policy information, and case information. The reply draft generation unit 460 may generate the reply draft within the user terminal 20, or it may send a generation request to the external generation processing device 40 and obtain the reply draft as a response. In one embodiment, the reply text generation unit 460 may generate a reply text in a first language corresponding to the language of the review content and a display text in a second language corresponding to the language set on the user terminal 20. The reply text generation unit 460 then proceeds to step S606.

[0104] (Step S606) The display control unit 490 displays the reply draft generated in step S605, and may also display display text, referenced case information, or progress information as needed. In one embodiment, if multiple reviews are managed according to a predetermined processing order, the display control unit 490 may sequentially display a draft reply for each review according to the processing order. The display control unit 490 then proceeds to step S607.

[0105] (Step S607) The confirmed reply acquisition unit 470 acquires the confirmed reply content based on the reply draft. For example, the confirmed reply acquisition unit 470 may acquire the content that was the subject of the confirmation operation by the user, the content that was finally entered in the reply input area, or the content that was finally selected as the target for transmission. The confirmed reply may be a reply that the user edited before confirming it, or it may be a reply that was confirmed without any editing. The confirmation reply acquisition unit 470 then proceeds to step S608.

[0106] In one embodiment, the confirmed reply acquisition unit 470 may identify the confirmed reply content as the recipient of transmission. Furthermore, the display control unit 490 or the confirmed reply acquisition unit 470 may automatically execute a transmission request for the confirmed reply content to the platform or an external system cooperating therewith when predetermined execution conditions are met.

[0107] (Step S608) The case registration control unit 480 controls the registration of the confirmed reply content as case information that can be used to generate subsequent reply drafts. For example, the case registration control unit 480 may send the review content and the confirmed reply content to the management server 30 and store it as case information in the management server 30. In one embodiment, the case registration control unit 480 may calculate the difference between the draft reply and the finalized reply, and store the finalized reply as case information only if the difference satisfies predetermined conditions. The difference may be calculated based on at least one of the following: character difference, editing distance, added text, deleted text, or replacement text. The predetermined conditions may include, for example, a threshold condition for the difference, a condition that the difference is within a predetermined range, a condition corresponding to a predetermined editing mode, or case registration conditions based on learning conditions, review type, evaluation category, usage frequency, period, legal risk information, or policy compliance information set for each store, business operator, or account. In one embodiment, the case registration control unit 480 may register only confirmed reply content corresponding to the language set as the target language for learning as case information. The case registration control unit 480 then proceeds to step S609.

[0108] (Step S609) The display control unit 490 or the review acquisition unit 420 determines whether there are any further reviews to be processed. If there are further reviews to be processed, the review acquisition unit 420 may set the next review as the target review according to a predetermined processing order such as in order of posting date and time, acquisition order, importance order, evaluation order, unaddressed period order, or user-specified order, and return to step S601. If there are no further reviews to be processed, the information processing system 1000 may terminate the series of processes.

[0109] Next, we will describe a method in which review content is obtained using a screen area specified by the user, a reply input area is identified, and a reply draft is provided for input. Figure 7 is an example of a flowchart relating to this method.

[0110] (Step S651) The user terminal 20 starts rectangular selection mode in response to a predetermined operation input. The predetermined operation input may be, for example, a shortcut key input, selection of an activation button on a toolbar, selection from a context menu, or a similar operation.

[0111] In one embodiment, the user terminal 20 may superimpose a semi-transparent overlay, a pointer indicating the start of selection, or a display indicating that rectangular selection is possible onto the display area in the web browser in response to the start of rectangular selection mode. Furthermore, if a previously selected screen area exists, the selection range corresponding to that previously selected screen area may be redisplayed.

[0112] (Step S652) The screen area analysis unit 495 acquires a screen area based on a user's drag operation, range selection operation, or similar input. The screen area may be specified to include both the review text and the reply field for that review, or it may be specified to primarily include the review text.

[0113] (Step S653) The screen area analysis unit 495 analyzes the text and element structure contained within the acquired screen area. For example, the screen area analysis unit 495 may analyze the display elements within the screen area by performing element analysis based on a document object model, extracting text nodes, acquiring display position information, or acquiring hierarchical structure information.

[0114] (Step S654) The screen area analysis unit 495 extracts candidate review texts and evaluation expression candidates based on the analysis results. For example, the screen area analysis unit 495 may assign scores to the acquired text nodes based on the number of characters, number of lines, position, line break structure, relationship with surrounding elements, proximity to evaluation expression, or proximity to author information, and determine candidate review texts based on these scores.

[0115] In one embodiment, the screen area analysis unit 495 may prioritize text groups adjacent to star ratings, numerical ratings, icon displays, or graded rating displays as candidates for the review text.

[0116] (Step S655) The screen area analysis unit 495 detects editable elements that exist within or near the screen area. Editable elements may include, for example, text areas, input boxes, content editable elements, input form elements, or similar elements.

[0117] (Step S656) The screen area analysis unit 495 may calculate proximity or relevance for each of the detected editable elements based on the vertical distance from the screen area, the amount of horizontal overlap, whether or not they are located below the screen area, the display state of the element, the editable state, the hierarchical relationship on the document object model, the label string, the placeholder string, the positional relationship with a button or heading, etc., and may identify the editable element with the highest proximity or relevance as the reply input area.

[0118] With this configuration, it is possible to estimate the correspondence between the review text and the reply input area based on the spatial proximity and element structure on the screen, without having to pre-maintain fixed element selectors for each platform.

[0119] (Step S657) The display control unit 490 may display the review text candidates, evaluation expression candidates, reply input area candidates, or parts thereof, extracted by the screen area analysis unit 495 in a confirmation area within a confirmation pop-up, dialog, side panel, or overlay.

[0120] In one embodiment, the display control unit 490 or the screen area analysis unit 495 may accept user confirmation. If user confirmation is accepted and the user does not accept it, the screen area analysis unit 495 may return to step S652 to accept reacquisition of the screen area, or may re-execute the candidate extraction or candidate identification from step S653 onwards.

[0121] On the other hand, if user confirmation is omitted, or if the user consents, the user terminal 20 may start the reply draft generation process by the reply draft generation unit 460 based on the review text candidate, evaluation expression candidate, platform type information, industry information, case information, and other input information.

[0122] (Step S659) The reply draft generation unit 460 generates a reply draft based on the review text candidate, evaluation expression candidate, platform type information, industry information, and case information. In an embodiment where user confirmation is accepted, the reply draft generation unit 460 may generate a reply draft based on the accepted review text candidate, evaluation expression candidate, platform type information, industry information, and case information. The reply draft generation unit 460 may generate the reply draft within the user terminal 20, or it may send a generation request to the external generation processing device 40 and obtain the reply draft as a response.

[0123] In one embodiment, the user terminal 20 may output an icon, animation, progress indicator, or waiting indicator to a predetermined location on the screen, for example, the lower right area of ​​the screen, in response to the start of the reply draft generation process.

[0124] (Step S660) The display control unit 490 or the confirmed reply acquisition unit 470 displays the generated reply draft or makes it available for input into the reply input area. For example, the display control unit 490 may display the reply draft in a confirmation pop-up, side panel, preview area or reply candidate display area, output the reply draft to the clipboard, provide input assistance to reflect the reply draft in the reply input area, or temporarily reflect the reply draft in the reply input area.

[0125] In one embodiment, the display control unit 490 or the confirmed reply acquisition unit 470 may reflect the generated reply draft or the confirmed reply content based thereon into a specified reply input area. In an embodiment that accepts user confirmation, the display control unit 490 or the confirmed reply acquisition unit 470 may reflect the generated reply draft or the confirmed reply content based thereon into a specified reply input area in response to the user's acceptance operation. The reflection into the reply input area may be performed by automatic input, paste assistance, input candidate presentation, or similar means.

[0126] In one embodiment, the operation to send the reply content may be performed by the user, and the user terminal 20 may be configured not to automatically press the send button or perform automatic sending. On the other hand, in environments where automatic sending is permitted by platform specifications, contract terms, permission settings, or operational rules, the display control unit 490 or the confirmed reply acquisition unit 470 may, after reflecting the generated reply draft or the confirmed reply content based thereon in the reply input area, identify it as a target for transmission, and automatically execute a transmission request. After that, the confirmed reply content based on the reply draft may be acquired, and the process may proceed to step S608 in Figure 6 as necessary.

[0127] In one embodiment, when the user terminal 20 provides continuous support for responding to multiple reviews, it may display ranking information for the currently targeted review, such as the current number of reviews relative to the total number of reviews. Furthermore, the user terminal 20 may display candidate response messages for the next review once the response message for the current review has been finalized or entered.

[0128] In one embodiment, the user terminal 20 may pre-acquire or generate a draft reply corresponding to the next review to be processed. This reduces the user's waiting time when replying to multiple reviews sequentially.

[0129] Thus, according to this embodiment, based on a screen area specified by a user-initiated rectangular drag operation, the extraction of review text candidates, estimation of the reply input area, user confirmation, reply draft generation, and reply input support can be realized as a series of workflows. Therefore, reply support can be adapted to various screen structures without depending on a fixed platform-specific implementation.

[0130] Note that the execution order of each step shown in Figure 7 is not limited to the example above. For example, steps S653 and S654 may be executed in parallel, or at least a portion of steps S653 to S656 may be re-executed depending on the verification result in step S657.

[0131] As shown in Figures 6 and 7, the process not only provides support for responding to reviews on the review management interface, but also generates draft responses based on control conditions and case information, and registers the finalized response content as a case study, for reviews or similar information obtained from screen areas specified by the user.

[0132] Note that the execution order of each step is not limited to the example above. For example, some control conditions may be obtained before obtaining the review content, and the acquisition of case information may be performed in parallel with the identification of reply policy information. In addition, some of the processing related to the generation of the reply text and the registration of case information may be divided and executed among the user terminal 20, the management server 30, and the external generation processing device 40.

[0133] (3) Effects of the embodiment According to this embodiment, in addition to the review content corresponding to the target review, a draft reply can be generated based on industry information, evaluation expression information, platform type information, other control conditions, and case information including past review content and corresponding reply content. This makes it possible to generate a draft reply that is not simply a response to the review text, but is tailored to the operational policies, reply trends, and usage environment of each store, business, or account.

[0134] Furthermore, according to this embodiment, the reply content finalized based on the reply draft can be stored as example information that can be used to generate subsequent reply drafts. Therefore, if a user edits the reply draft, the editing results can be reflected, and even if the reply draft is finalized without editing, the finalized result can be used to generate subsequent reply drafts. As a result, as the system is used more and more, it becomes possible to continuously generate reply drafts that take into account the actual reply history of each store, business, or account.

[0135] Furthermore, according to this embodiment, it is possible to flexibly respond to the types of evaluation expressions, additional condition information, language conditions, sequential handling of multiple reviews, and input support based on the specification of screen areas, so it can be applied to various platforms or various screen configurations, and it is possible to improve the efficiency of review response work, improve the quality of responses, and continuously optimize operations.

[0136] (modified version) Hereinafter, Modifications 1 to 10 are described as examples of how the configurations and processes according to the embodiment can be implemented in various modified forms. The configurations or processes described in each of the following modifications can be combined in any way as long as they do not contradict each other. Furthermore, the configurations or processes described in each modification are arbitrary configurations that can be additionally applied to the basic process of acquiring review content, acquiring control conditions, identifying reply policy information, acquiring case information, generating a reply draft, acquiring confirmed reply content, and storing the confirmed reply content as case information, and the basic process can be executed even if the modifications are not applied.

[0137] (Modification 1: Division of responsibilities for executing the process) The division of processing among the user terminal 20, the management server 30, and the external generation processing device 40 is not limited to the above-described configuration. For example, the user terminal 20 may perform the acquisition of review content, acquisition of control conditions, identification of reply policy information, and acquisition of confirmed reply content, while the management server 30 may only perform the storage and provision of case information. Alternatively, the management server 30 may perform the generation of the reply text, or the external generation processing device 40 may perform most of the reply text generation. Furthermore, the user terminal 20 and the management server 30 may jointly generate the reply text.

[0138] This configuration allows for flexible adjustment of processing load depending on the client's computing resources, the server's computing resources, communication load, or the usage environment.

[0139] (Modification 2: Modification of the source from which the target input information is obtained) The source of the target input information is not limited to reviews displayed on the review management interface. For example, the target input information may be obtained from information displayed on a web browser, from locally stored cached information, or from input information stored on the management server 30. Furthermore, the target input information may also be reviews, inquiries, reservation requests, free-response answers in surveys, message bodies, or other user input information.

[0140] This configuration allows for the application of the system not only to review screens but also to generating response texts for various user inputs, case learning, and subsequent processing.

[0141] (Modification 3: Expansion of the configuration of control conditions) Control conditions are not limited to industry information, evaluation expression information, and platform type information. For example, control conditions may include time of day information, day of the week information, seasonal information, store type information, service type information, staff attribute information, membership type information, response priority information, or campaign information. Furthermore, some control conditions may be provided by user selection input or automatically extracted from review content or surrounding information.

[0142] This configuration allows for increasing or decreasing the number of criteria used to identify reply policy information according to actual operational practices, thereby improving the suitability of the reply draft.

[0143] (Modification 4: Variation of the method for identifying reply policy information) The method for identifying response policy information is not limited to a fixed rule or matrix-based approach. For example, response policy information may be identified by table referencing based on a combination of multiple condition axes, based on a prioritized rule, or estimated by a trained model. Alternatively, multiple candidate response policy information may be generated and one selected from among them, or multiple response policy information may be combined into a single policy.

[0144] This configuration allows for gradual or adaptive changes to the logic for determining reply policy information, thereby improving the quality and consistency of the reply text.

[0145] (Modification 5: Variation of the method for obtaining case information) The case information acquisition unit 450 may acquire case information in stages. For example, as the first stage, past cases corresponding to the same store or account may be acquired, and if the number of cases or similarity does not meet the requirements, as the second stage, cases belonging to the same industry or platform may be acquired supplementarily. Furthermore, the acquisition of case information may be performed by keyword matching, classification information matching, similarity determination based on embedding vectors, or a combination thereof.

[0146] This configuration allows for prioritizing recent examples first, and only expanding the acquisition range if necessary, thus achieving both stability and computational efficiency in generating reply templates.

[0147] (Variation 6: Obtaining confirmed reply content and modifying registration conditions) Confirmed replies may be obtained based on an explicit confirmation operation by the user, based on the elapsed time after they are reflected in the reply input area, or based on their selection as a target for transmission. Furthermore, the registration conditions for case information are not limited to the difference amount meeting a predetermined condition, but may also include reply length, review type, review importance, explicit learning permission operation by the user, or not falling under a predetermined prohibition condition.

[0148] This configuration allows for adjustment of the triggers or conditions for registering cases depending on the operational style, enabling the accumulation of useful response data while suppressing the inclusion of unnecessary training data.

[0149] (Variation 7: Variation of language processing) The language processing method is not limited to bilingual output. For example, the reply draft generation unit 460 may generate a draft corresponding to a third language different from the language of the review content, the language set on the user terminal 20, and the target language for learning. Furthermore, the display draft is not limited to a word-for-word translation, but may also be a summary, an explanatory text, or a confirmation draft. Moreover, the items stored as case information may not only be the finalized reply content corresponding to the target language for learning, but also the translated draft, the normalized draft, or the summary.

[0150] This configuration allows for improved operational flexibility in multilingual environments.

[0151] (Variation 8: Variation of the multiple review process) Processing multiple reviews is not limited to simple sequential generation. For example, a response draft may be generated sequentially for each review, or response drafts for multiple reviews may be generated in advance. Alternatively, the next candidate response draft may be obtained in advance, and the next review may be displayed in response to the user's confirmation action. Furthermore, the order in which reviews are processed may be by acquisition order, posting date and time, importance order, rating order, unanswered period order, or user-specified order.

[0152] This configuration reduces waiting times during multiple review processes and improves usability during continuous processing.

[0153] (Variation 9: Modification of input support mode) The method of providing a reply draft for input into the reply input area is not limited to simply displaying it. For example, the reply draft may be output to the clipboard, automatically reflected in the reply input area, or displayed in a format where each candidate sentence can be selected. Furthermore, the identification of editable elements adjacent to the screen area is not limited to methods based on their positional relationship on the screen, but may also be based on hierarchical relationships in the document object model, element attributes, label information, or a combination thereof.

[0154] This configuration allows for changing the input support method depending on differences in screen structure or operation method.

[0155] (Modification 10: Variation of subsequent processing after confirmation) Subsequent processing of confirmed replies is not limited to storage as case information. For example, confirmed replies may only be reflected in the reply input area, or they may be sent to the platform in response to a user's submission operation, or in a given environment, they may be automatically passed to subsequent processing as a target for transmission. Furthermore, necessary information related to confirmed replies may be linked to reservation management systems, inquiry management systems, customer support systems, or business support systems. In addition, subsequent processing may be executed automatically or conditional on a user's confirmation operation.

[0156] This configuration allows the reply draft generation and learning process to be connected not only to review replies but also to various customer support or business processes, and enables the use of either automatic sending or user-activated sending depending on the platform's specifications or operating conditions.

[0157] Furthermore, the configurations described in the above modified examples do not limit the technical scope of the invention as described in the claims, and can be appropriately selected or combined depending on the embodiment.

[0158] <Note> This embodiment includes the following disclosures.

[0159] (Note 1) An information processing system, Having one or more control units, The control unit, Retrieve the review content corresponding to the target review, Obtain one or more control conditions related to the aforementioned review, Identify the reply policy information corresponding to the aforementioned control conditions, From the case memory unit associated with the store, business, or account, case information including past review content related to the aforementioned review content and corresponding reply content is obtained. Based on the review content, the response policy information, and the case information, a draft response is generated. Based on the aforementioned draft reply, obtain the finalized reply content, The confirmed reply content is stored as example information that can be used to generate subsequent reply drafts. Information processing system.

[0160] According to this addendum, a draft reply can be generated based on the review content, control conditions, and case information, and the finalized reply can be reused in subsequent instances, thereby improving the suitability and continuous improvement of reply generation.

[0161] (Note 2) The information processing system described in Appendix 1, The one or more control conditions described above are: This includes industry information, evaluation expression information, and platform type information. Information processing system.

[0162] This addendum allows for switching response policies according to industry, evaluation style, and platform type, making it easier to generate response templates that reflect actual operational practices.

[0163] (Note 3) The information processing system described in Appendix 2, The one or more control conditions described above are: In addition to the aforementioned industry information, evaluation expression information, and platform type information, the following additional condition information corresponding to the additional axis is included: Information processing system.

[0164] According to this addendum, additional conditions can be considered in addition to the basic three-axis conditions, thereby improving the flexibility of reply text generation.

[0165] (Note 4) The information processing system described in Appendix 3, The aforementioned additional conditional information is, Includes at least one of the following: review language information, user language information, regional information, legal risk information, or policy compliance information. Information processing system.

[0166] This addendum allows for the adoption of response policies that take into account language, regional, or risk conditions, thereby enhancing multilingual support and operational security.

[0167] (Note 5) An information processing system as described in Appendix 2 or Appendix 3, The aforementioned evaluation expression information is, Includes at least one of the following: star rating, icon count, numerical rating, or graded rating. Information processing system.

[0168] This addendum allows the application to various services or screen configurations with different evaluation formats, thereby expanding the scope of application.

[0169] (Note 6) An information processing system described in any one of the appendices 1 to 5, The control unit, As the aforementioned case information, we obtain past review content similar to the aforementioned review content and the actual replies sent to that past review content. Information processing system.

[0170] According to this note, response templates can be generated by referencing similar past reviews and actual responses sent, making it easier to reflect response trends specific to each store, business, or account.

[0171] (Note 7) An information processing system described in any one of the appendices 1 to 6, The control unit, If the difference between the draft reply and the finalized reply satisfies a predetermined condition, the finalized reply is stored as the case information. Information processing system.

[0172] According to this note, significant and definitive results where the difference satisfies predetermined conditions can be selected and used as training data, thus allowing for the reflection of useful cases while suppressing the accumulation of unnecessary cases.

[0173] (Note 8) The information processing system described in Appendix 7, The aforementioned difference amount is It is calculated based on at least one of the following: character difference, edit distance, amount of added text, amount of deleted text, or amount of replacement. Information processing system.

[0174] According to this addendum, the difference can be calculated from multiple perspectives, making it easier to appropriately set the case registration conditions according to actual operational conditions.

[0175] (Note 9) An information processing system described in any one of the appendices 1 to 8, The control unit, A draft reply in the first language corresponding to the language of the aforementioned review content, A second language display text corresponding to the language set on the user's terminal, and Output Information processing system.

[0176] According to this addendum, a draft response tailored to the review content and a user-friendly display version can be provided together, making verification and operation easier in multilingual environments.

[0177] (Note 10) The information processing system described in Appendix 9, The control unit, Only the confirmed reply content corresponding to the language set as the target language for learning is stored as the case information. Information processing system.

[0178] According to this note, it becomes possible to control the languages ​​to be learned, making it easier to suppress the instability of learning caused by language mixing.

[0179] (Note 11) An information processing system described in any one of the appendices 1 to 10, The control unit, When multiple reviews are managed according to a predetermined processing order, the aforementioned draft response is generated sequentially for each review according to that processing order. Information processing system.

[0180] According to this note, it is possible to generate response drafts for multiple reviews consecutively, thereby improving processing efficiency when handling multiple items.

[0181] (Note 12) An information processing system described in any one of the appendices 1 through 11, The control unit, The review content is obtained from within the screen area specified by the user. An editable element adjacent to the aforementioned screen area is identified as the reply input area. The aforementioned reply draft is provided for input into the reply input area. Information processing system.

[0182] According to this addendum, response support can be provided not only to fixed review management screens, but also to reviews or similar information displayed on any screen, thus expanding the range of applicable screen configurations.

[0183] (Note 13) An information processing system described in any one of the appendices 1 to 12, The control unit, The confirmed reply content is identified as the recipient, When predetermined execution conditions are met, the system automatically executes a request to send the confirmed reply content to the platform or an external system that cooperates with it. Information processing system.

[0184] According to this addendum, the confirmed reply content can be identified as the recipient, and the sending request can be automatically executed when the predetermined execution conditions are met. Therefore, the sending process after the reply draft is generated can be efficiently executed according to the platform specifications, contract terms, permission settings, or operational rules.

[0185] (Note 14) An information processing method performed by an information processing system, Retrieve the review content corresponding to the target review, Obtain one or more control conditions related to the aforementioned review, Identify the reply policy information corresponding to the aforementioned control conditions, From the case memory unit associated with the store, business, or account, case information including past review content related to the aforementioned review content and corresponding reply content is obtained. Based on the review content, the response policy information, and the case information, a draft response is generated. Based on the aforementioned draft reply, obtain the finalized reply content, The confirmed reply content is stored as example information that can be used to generate subsequent reply drafts. Information processing methods.

[0186] According to this addendum, it becomes possible to understand the entire information processing procedure related to generating reply drafts and registering cases, rather than being limited to the configuration of the device, thus making it easier to grasp the invention from a processing perspective.

[0187] (Note 15) It is a program, Computers, A program to function as an information processing system as described in any one of the appendices 1 through 13.

[0188] According to this appendix, each of the configurations described in Appendix 1 to Appendix 12 can be implemented on a computer, and therefore can serve as the basis for patenting corresponding to a program product or software distribution format. [Explanation of symbols]

[0189] 20 user terminals 30 Management Servers 40 External generation processing device 50 Networks 210 Control Unit 220 Storage section 230 Input section 240 Output section 250 Communications Department 260 Internal Bus 1000 Information Processing Systems

Claims

1. An information processing system, Having one or more control units, The control unit, Retrieve the review content corresponding to the target review, Obtain one or more control conditions related to the aforementioned review, Identify the reply policy information corresponding to the aforementioned control conditions, From the case memory unit associated with the store, business, or account, case information including past review content related to the aforementioned review content and corresponding reply content is obtained. By providing the aforementioned review content, the aforementioned reply policy information, and the aforementioned case information as input information for generation to an external generation processing device or language model, a draft reply is generated using the external generation processing device or language model. The confirmed reply content is obtained based on the user's confirmation, editing, or confirmation operation on the aforementioned reply draft. The difference between the aforementioned draft reply and the previously confirmed reply is calculated, If the difference amount satisfies predetermined conditions, the information including the review content and the finalized reply content is stored as example information that can be used to generate subsequent reply drafts. Information processing system.

2. The information processing system according to claim 1, The one or more control conditions are: This includes industry information, evaluation expression information, and platform type information. Information processing system.

3. The information processing system according to claim 2, The one or more control conditions are: In addition to the aforementioned industry information, evaluation expression information, and platform type information, the following additional condition information corresponding to the additional axis is included: Information processing system.

4. The information processing system according to claim 3, The aforementioned additional condition information is, Includes at least one of the following: review language information, user language information, regional information, legal risk information, or policy compliance information. Information processing system.

5. The information processing system according to claim 2, The aforementioned evaluation expression information is, Includes at least one of the following: star rating, icon count, numerical rating, or graded rating. Information processing system.

6. The information processing system according to claim 1, The control unit, As the aforementioned case information, we obtain past review content similar to the aforementioned review content and the actual replies sent to that past review content. Information processing system.

7. The information processing system according to claim 1, The predetermined conditions include at least one of the following: the difference amount is greater than or equal to a predetermined threshold; the difference amount is less than or equal to a predetermined threshold; the difference amount is within a predetermined range; or the difference amount corresponds to a predetermined editing mode. Information processing system.

8. The information processing system according to claim 1, The aforementioned difference amount is It is calculated based on at least one of the following: character difference, edit distance, amount of added text, amount of deleted text, or amount of replacement. Information processing system.

9. The information processing system according to claim 1, The control unit, A draft reply in the first language corresponding to the language of the aforementioned review content, A second language display text corresponding to the language set on the user's terminal, and Output Information processing system.

10. The information processing system according to claim 9, The control unit, Only the confirmed reply content corresponding to the language set as the target language for learning is stored as the case information. Information processing system.

11. The information processing system according to claim 1, The control unit, When multiple reviews are managed according to a predetermined processing order, the aforementioned draft response is generated sequentially for each review according to that processing order. Information processing system.

12. The information processing system according to claim 1, The control unit, The review content is obtained from within the screen area specified by the user. An editable element adjacent to the aforementioned screen area is identified as the reply input area. The aforementioned reply draft is provided for input into the reply input area. Information processing system.

13. The information processing system according to claim 1, The control unit, The confirmed reply content is identified as the recipient, When predetermined execution conditions are met, the system automatically executes a request to send the confirmed reply content to the platform or an external system that cooperates with it. Information processing system.

14. An information processing method performed by an information processing system, Retrieve the review content corresponding to the target review, Obtain one or more control conditions related to the aforementioned review, Identify the reply policy information corresponding to the aforementioned control conditions, From the case memory unit associated with the store, business, or account, case information including past review content related to the aforementioned review content and corresponding reply content is obtained. By providing the aforementioned review content, the aforementioned reply policy information, and the aforementioned case information as input information for generation to an external generation processing device or language model, a draft reply is generated using the external generation processing device or language model. The confirmed reply content is obtained based on the user's confirmation, editing, or confirmation operation on the aforementioned reply draft. The difference between the aforementioned draft reply and the previously confirmed reply is calculated, If the difference amount satisfies predetermined conditions, the information including the review content and the finalized reply content is stored as example information that can be used to generate subsequent reply drafts. Information processing methods.

15. Computers, A program for functioning as an information processing system according to any one of claims 1 to 13.

Images