Server device, control method, and program

Abstract

To provide a technique that improves convenience in retrieving customer support examples.SOLUTION: A server device, which provides a chat service for retrieving customer support examples, provides a user interface to a client device, stores customer support examples and past inquiries, presents frequently occurring character strings included in the past inquiries via the user interface, accepts input from the client device via the user interface for an inquiry about a customer support example, obtains past support example data highly relevant to the inquiry upon receiving a response request for a support example, and presents at least a part of the obtained support example data via the user interface.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a server device, a control method, and a program. [Background technology]

[0002] Various services have been proposed to accept consultations via chat over a network. Patent Document 1 discloses an online legal consultation system that searches for and provides legal consultation data via chat. Patent Document 2 discloses a chatbot that chats with users using a large-scale language model. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-141223 [Patent Document 2] Japanese Patent Publication No. 2022-176922 Summary of the Invention [Problem to be solved by the invention]

[0004] Here, when searching for past cases of customer service, it has been an issue to reduce the input work required by the user making an inquiry.

[0005] The present invention has been made in view of the above-mentioned problems, and aims to provide a technique for improving convenience in searching for case studies for customers. [Means for solving the problem]

[0006] In order to solve the above problem, a server device of the present invention comprises: A server device that provides a chat service for searching for cases of responses to customers, providing means for providing a user interface to the client device; a storage means for storing case studies of responses to customers and past inquiries; a first presentation means for presenting, via the user interface, character strings that are frequently included in the past queries; a reception unit that receives an input of an inquiry about a case example from a customer from the client device via the user interface; an acquisition means for acquiring past case data highly relevant to the query in response to the reception means receiving a search instruction for case data; a second presentation means for presenting at least a part of the corresponding case data acquired by the acquisition means via the user interface; The present invention is characterized by comprising: [Effects of the Invention]

[0007] According to the present invention, it is possible to provide a technique that improves convenience in searching for case studies for customers. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a configuration diagram of a chat system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram showing the hardware configuration of a server device according to the embodiment. [Figure 3] FIG. 2 is a diagram showing the software configuration of a server device according to the embodiment. [Figure 4] 1A is a diagram showing an example of user information according to the present embodiment, FIG. 1B is a diagram showing an example of the structure of corresponding case data, and FIG. 1C is a diagram showing an example of the structure of inquiry data. [Figure 5] 10A to 10C are diagrams showing examples of chat screen displays. [Figure 6] FIG. 4 is a diagram showing an example of processing executed by the server device according to the embodiment. [Figure 7] 10A and 10B are diagrams showing another example of the display of the chat screen. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, the embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the invention as claimed, and not all combinations of features described in the embodiments are necessarily essential to the invention. Two or more of the features described in the embodiments may be combined in any desired manner. Furthermore, the same reference numerals are used to designate identical or similar components, and redundant descriptions will be omitted.

[0010] First Embodiment A case study search system operated by a server device according to this embodiment will be described with reference to Figure 1. The server device 10 according to this embodiment provides a chat service and, in response to an inquiry from a client device 20 via the chat service, provides information about past cases of dealing with customers. In this embodiment, staff provide a service or product to a customer, and cases of dealing with the customer during the process of providing the service or product are stored in the server device, and the stored cases are searched to provide the cases to the staff, allowing the staff to improve their customer service or determine a course of action for dealing with the customer.

[0011] Here, the staff will be described as employees working at any facility such as a specific store, hospital, government office, school, etc., but may also be individuals and is not limited to this. Furthermore, the customers will be described as including patients and students who receive services or products at the above-mentioned facilities, but are not limited to this. For example, the customers may also be animals other than humans.

[0012] The staff operates the client device 20 to inquire about case studies from the server device 10. In response to the inquiry, the server device 10 provides information about case studies to the client device 20. Note that, although the example in FIG. 1 shows that the server device 10 and the client device 20 are directly connected, they may communicate via at least one of the Internet and an intranet.

[0013] <Hardware configuration> As shown in FIG. 2A, the server device 10 includes a processor 11, a memory 12, a storage 13, and a network interface (IF) 14.

[0014] The processor 11 is a system control unit that controls the entire server device 10. The processor 11 includes at least one of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and a microcontroller that executes a program or an operating system (OS) stored in the storage 13.

[0015] The memory 12 is composed of a dynamic random access memory (DRAM) or a static random access memory (SRAM) used as a work memory for the processor 11. The storage 13 includes a hard disk drive (HDD) or a solid state drive (SSD) that stores the programs executed by the processor 11, the OS, and at least one of data tables and setting values ​​used by the programs. The storage 13 may also include a programmable read only memory (PROM), an erasable read only memory (EPROM), or an electrically erasable read only memory (EEPROM). The storage 13 stores the contents of inquiries previously received from the client device 20 and response examples.

[0016] Although the server device 10 will be described as a single device, multiple devices may work together to realize the functions of the server device 10. For example, the storage 13 may be an external storage medium, a storage device capable of communicating with the server device 10, or cloud storage located on a network. Furthermore, the server device 10 may be a virtual PC implemented on a cloud network.

[0017] As shown in FIG. 2B, the client device 20 includes a processor 21, a memory 22, a storage 23, a network IF 24, an input / output IF 25, and a display 26.

[0018] The processor 21 is a system control unit that controls the entire client device 20. The processor 21 includes at least one of a CPU, a GPU, and a microcontroller that executes a program or an OS stored in the storage 23.

[0019] The memory 22 is composed of a DRAM, an SRAM, etc. used as a work memory for the processor 21. The storage 23 includes an HDD or an SSD that stores the programs executed by the processor 21, the OS, and at least one of data tables and setting values ​​used by the programs. The storage 23 may also include a PROM, an EPROM, or an EEPROM.

[0020] The network IF 24 is an interface for connecting to an external device (not shown) and performing data communication, and includes at least one of a wired network IF and a wireless network IF. The wireless network IF can perform wireless communication in accordance with the 3GPP (registered trademark) standard, Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] The input / output IF 25 includes at least one of an input unit such as a keyboard, mouse, buttons, and microphone for receiving user operations and an output unit such as a speaker, or includes an interface connectable to an input device and an output device.

[0022] The display 26 provides the user with information about the corresponding case. In one example, the input / output IF 25 and the display 26 may be integrally formed as a touch panel display.

[0023] 3(A) shows a functional configuration diagram of the server device 10. The processor 11 of the server device 10 executes programs stored in the storage 13, thereby functioning as a user interface (UI) providing unit 201, an inquiry receiving unit 202, a response generating unit 203, a response providing unit 204, a case management unit 205, an inquiry management unit 206, an input candidate presenting unit 207, an emergency contact recommending unit 208, a case learning unit 209, a user information storage unit 210, a case storage unit 211, and a query storage unit 212.

[0024] The UI providing unit 201 provides a user interface for receiving an inquiry from the client device 20 and providing information about corresponding cases to the client device 20. In one example, the UI providing unit 201 operates as a web server and receives an inquiry via a web browser of the client device 20.

[0025] The query receiving unit 202 receives a query in the form of text, image, video, or voice from the client device 20 via the UI provided by the UI providing unit 201. As will be described later, the query receiving unit 202 acquires the query being input at a predetermined timing. The predetermined timing may be when a predetermined time, such as 5 or 10 seconds, has elapsed, or when a new character string is input in the query.

[0026] The response generation unit 203 analyzes a query received via the UI provided by the UI provision unit 201, identifies highly relevant corresponding case data from the corresponding case data stored in the case storage unit 211, and generates a response. In one example, the response generation unit 203 may include a generative AI (Artificial Intelligence) such as a large-scale language model, and may use the corresponding case data for learning to generate a response based on at least a portion of the corresponding case data during text-based interactions with the user of the client device 20. In another example, the response generation unit may analyze the query and transmit at least a portion of the highly relevant corresponding case data identified as a response. The generative AI includes a general-purpose natural language processing learning model such as a large-scale language model (LLM) that has learned from a huge amount of data.

[0027] The response providing unit 204 provides the response generated by the response generating unit 203 to the client device 20 via the UI. The form of the response provided by the response providing unit 204 will be described later with reference to Figures 5(C), 6(A), and 6(B).

[0028] The case management unit 205 adds corresponding cases to the case storage unit 211 and records progress data on cases for which advice was provided in response to inquiries in the past. The inquiry management unit 206 records inquiries received via the UI in the inquiry storage unit 212. In one example, the inquiry management unit 206 may tally up words included in inquiries received via the UI, or may identify and store categories that are highly relevant to inquiries received via the UI.

[0029] The input candidate presentation unit 207 acquires the query being input as text from the client device 20 via the UI provided by the UI provision unit 201, and presents candidate character strings that are likely to be input with the same query as the query being input. The candidate terms that are likely to be input include words that are the same as or highly related to words included in the query being input, and can be identified by identifying query data stored in the query storage unit 212 and analyzing words that are frequently included in the query data. Alternatively, the candidate terms that are likely to be input include words that are the same as or highly related to words included in the query being input, and can be identified by identifying consultation content in case data stored in the case storage unit 211 and analyzing words that are frequently included in the consultation content. The input candidate presentation unit 207 may present candidate character strings that are likely to be input even when a query has not been input from the client device 20. For example, the input candidate presentation unit 207 may present words that are frequently included in queries previously received via the UI, categories that are highly related to queries previously received via the UI, or words that are frequently included in consultation content in case data stored in the case storage unit 211.

[0030] If the inquiry received via the UI is of a high urgency, i.e., an inquiry that requires immediate contact with an expert or the police, the emergency contact recommendation unit 208 recommends to the user of the client device 20 via the UI that they contact, consult with, or report to an expert or the police.

[0031] The case learning unit 209 is configured to indicate corresponding cases that are highly related to a corresponding case from the corresponding case data stored in the case storage unit 211. In one example, a machine learning model such as a large-scale language model is trained using the cases so that a query is input and at least a portion of the corresponding cases is output.

[0032] The user information storage unit 210 stores information about users. FIG. 4A shows an example of the structure of user information according to this embodiment. The user information includes a user ID 401, a user name 402, a name 403, an organization 404, contact information 405, an authority 406, and a password 407. The user ID 401 is a user identifier. The organization 404 registers the organization to which the client belongs or the location of the organization. The authority 406 is a field for setting whether the user is permitted to manage case studies. For example, if the authority 406 is set to "administrator," the user can edit, delete, add, and view the case studies stored in the case storage unit 211. On the other hand, if the authority 406 is set to "viewer (non-administrator)," the user may only be able to view the case studies stored in the case storage unit 211. Note that the authority setting is merely an example, and any other authority can be set, such as one that only allows editing, deleting, and adding case studies.

[0033] The case storage unit 211 stores information about handling cases (handling case data). Fig. 4(B) shows an example of the structure of handling case data according to this embodiment. The handling case data includes a case ID 411, consultation date and time 412, summary 413, category 414, consultation content 415, advice 416, and result 417.

[0034] The case ID 411 is an identifier of the case. The summary 413 is a summary or title of the consultation content, and can be set by a user who adds the case to the case storage unit 211.

[0035] The category 414 indicates which of the predefined categories of consultation the consultation falls into, and can be set by the administrator user. Examples of categories include terms such as "how to deal with a difficult patient," "online posting," "non-payment for medical treatment," and "claim for compensation," but categories can be set arbitrarily depending on the type of case managed by the server device 10. Categories can be added or deleted by the administrator user.

[0036] Consultation content 415 indicates the consultation content (consultation data) received from the client. Advice 416 indicates the content of advice (advice data) provided by a specific user such as an expert or by the server device 10 in response to the consultation. Result 417 indicates what changes have occurred in response to the advice. For example, result 417 stores information on whether an improvement in the situation has been observed or whether the problem has been solved.

[0037] For example, one application of this embodiment provides a service in which a specific user, such as an expert, accepts consultations by telephone or email and gives advice, and the server device 10 can be used to allow other users to view the cases accepted by the expert. In such a case, the specific user can register the consultation content 415, advice 416 on the consultation content, and results 417 showing changes due to the advice.

[0038] In another example, one case may be edited by multiple users, such as a client registering consultation content 415 via client device 20, a specific user such as an expert registering only advice on the registered consultation content, and the client viewing the advice 416 reporting the result 417.

[0039] The inquiry storage unit 212 stores a history of inquiries (inquiry data) for information related to case studies. An example of the structure of the inquiry data is shown in Fig. 4(C). The inquiry data includes an inquiry ID 421, an inquiry date and time 422, and an inquiry content 423.

[0040] The inquiry ID 421 is an identifier of the inquiry of the corresponding case, the inquiry date and time 422 is data indicating the date and time when the inquiry was received via the UI, and the inquiry content 423 indicates the inquiry input by the user.

[0041] 3B shows a functional block diagram of the client device 20. The client device 20 includes a display control unit 251, an input receiving unit 252, and a response receiving unit 253.

[0042] The display control unit 251 displays a UI provided by the server device 10 on the display 26. The input accepting unit 252 accepts an inquiry input from a user via the input / output IF 25. The response receiving unit 253 receives a response to the inquiry from the server device 10 via the network IF 24. As described above, the response includes at least a portion of the corresponding case.

[0043] Next, the UI provided by the UI providing unit 201 according to this embodiment will be described with reference to FIGS. 5(A) to 5(C).

[0044] 5(A) shows an initial screen of the UI provided by the UI providing unit 201. The initial screen 500 includes an input area 501 for inputting a query, a response request button 502 for requesting the generation of a response in response to the query, an input candidate presentation area 503, and a message area 504 for displaying one or more messages.

[0045] The input area 501 is an input window that, when selected by the user, allows input of a query via keyboard or voice. Images and videos may be pasted into the input area 501. The response request button 502 is a button that requests information on corresponding cases based on the content of the query entered in the input area 501. In one example, with text entered in the input area 501, information on corresponding cases may be requested by pressing a specific key such as the enter button on the keyboard, in which case the response request button 502 may not be displayed.

[0046] The input candidate presentation area 503 displays information that assists the user in entering a query into the input area 501. For example, the initial screen 500 may display character strings that are frequently used among past queries. Alternatively, categories of solution cases presented in response to past queries may be displayed. In this case, solution cases presented in response to past queries may be categorized, and the category that has been presented most frequently may be displayed. Furthermore, a character string displayed in the input candidate presentation area 503 is input into the input area 501 by a user's selection. This allows the user to efficiently create a query using the displayed input candidates. In one example, when a user selects an input candidate displayed in the input candidate presentation area 503, the server device 10 may determine that the input of the input candidate and a response request have been accepted.

[0047] The message area 504 displays one or more messages exchanged between the server device 10 and the client device 20. When a query or a response is sent, a message relating to the query or response is displayed.

[0048] FIG. 5B shows an example of a screen during query input. In the screen 510 of FIG. 5B, "the patient" is entered in the input area 511 as the beginning of the query. In this case, the input candidate presentation unit 207 updates the character string displayed in the input candidate presentation area 512 based on the content of the query being input. For example, queries that include the query being input may be identified from past queries stored in the query storage unit 212, and a character string that is frequently used among queries that include the query being input may be displayed in the input candidate presentation area 512. In another example, a case that includes the character string of the query being input may be identified from the case storage unit 211, and a character string that is frequently used among the case that includes the consultation content 415 that includes the character string of the query being input may be displayed in the input candidate presentation area 512. In another example, a case that includes the character string of the query being input may be identified from the case storage unit 211, and a category with a large number of categories among the case that includes the character string of the query being input may be displayed in the input candidate presentation area 512. The input candidates displayed in the input candidate presentation area 512 can also be added to the input area 511 in response to a user's selection, similar to the input candidate presentation area 503 in FIG. 5(A).

[0049] 5(C) is an example of a screen that is displayed when a response is requested by pressing response request button 502 after inputting a query. In screen 520 of FIG. 5(C), query message 521 indicating the content of the query and response message 522 to the query are displayed in message area 504. As described above, query message 521 indicates the content of the query that was entered in input area 511 and included in the message that was sent to server device 10 when response request button 502 was pressed. Response message 522 indicates the content of the response that was input to case learning unit 209, which learned using corresponding cases stored in case storage unit 211 of server device 10, and that was generated by response generation unit 203 using the output.

[0050] The response generation unit 203 creates responses using a generation AI, allowing users of client devices to inquire about cases through natural interaction. Also, by presenting input candidates based on the query being entered, the convenience of the user entering a query can be improved.

[0051] Fig. 6 shows an example of processing executed by the server device 10 according to this embodiment. The processing shown in Fig. 6 is realized by the processor 11 of the server device 10 executing a program stored in the storage 13. The processing shown in Fig. 6 is executed when a UI display request is received from the client device 20. In one example, the UI display request from the client device 20 is determined to have been received when the server device 10, which operates as a web server, accesses a web page from the client device 20.

[0052] In S601, the server device 10 acquires character strings that are frequently included in past inquiries.

[0053] In S602, the server device 10 generates an initial screen including character strings that are frequently included in past inquiries, and provides the initial screen to the client device 20.

[0054] In S603, the server device 10 acquires a query being input from the client device 20 via the UI, if any. The query being input is a character string entered in the input field 511 of FIG. 5(A). Upon acquiring the query being input, the server device 10 identifies a character string highly relevant to the query being input in S604. As described above, in S604, a character string that includes the character string of the query being input and that is likely to be included in past queries stored in the query storage unit 212 and past consultation contents stored in the case storage unit 211 can be determined to be a highly relevant character string. Alternatively, a category of a past consultation content that includes the character string of the query being input and that is stored in the case storage unit 211 may be determined to be a highly relevant character string.

[0055] In S605, the server device 10 presents terms highly relevant to the query being input. In S605, the server device 10 updates the input candidate presentation area 503 so that terms highly relevant to the query being input are displayed.

[0056] In S606, the server device 10 determines whether a response request for a corresponding case has been received. For example, it can be determined that a response request for a corresponding case has been received when the response request button 502 in FIG. 5(A) is pressed. If a response request for a corresponding case has been received (Yes in S606), the server device 10 proceeds to S607 and generates a response based on the query. If it has been determined that a response request for a corresponding case has not been received (No in S606), the server device 10 returns to S603 and continues to receive query input.

[0057] In S607, the server device 10 refers to the response generation unit 203 and inputs the query into the large-scale language model to identify highly relevant corresponding cases, or creates a response based on highly relevant corresponding cases, as described above.

[0058] Then, in S608, the server device 10 provides the generated response to the client device 20, and the processing of FIG. 6 ends.

[0059] As described above, the server device 10 according to this embodiment accepts an input of a query about a case study via a user interface, and generates and provides a response based on past case study data that is highly relevant to the query. This allows the user to efficiently search for past case studies, improving user convenience.

[0060] Furthermore, the server device 10 according to the present embodiment stores past queries and presents frequently occurring character strings included in the past queries as input candidates on the user interface, thereby enabling a user entering a query to efficiently enter a query based on the input candidates.

[0061] <Modification> In Figure 5(C), we have explained that the query content is input to a large-scale language model trained using past correspondence cases, and the output is displayed as a response on the UI. However, it is also possible to generate a response based on correspondence cases, or to determine whether to provide advice based on case data depending on the query content.

[0062] 7(A) and 7(B) show modified examples of the response.

[0063] FIG. 7(A) shows an example of a response displaying a uniform resource locator (URL) at which corresponding cases highly relevant to the query can be displayed. In the response of FIG. 7(A), a message 700 including the summary of FIG. 4(B) and a URL at which the corresponding cases can be displayed is displayed in the message area. The user can check the corresponding cases by accessing the URL via a UI displayed on the client device 20. For example, the server device 10 can provide the URL of a web page generated based on the corresponding cases.

[0064] As a result, the server device 10 only needs to identify corresponding cases that are highly relevant to the query, so there is no need to generate a response using a large-scale language model or the like, and a classifier or the like can be used, thereby improving the design freedom of the response generation unit 203.

[0065] FIG. 7B shows an example of a display that determines the urgency of an inquiry and, depending on the urgency, urges the user to contact a specialist or a designated institution such as the police or a hospital, rather than providing a case for response.

[0066] In this embodiment, a response to a query is generated using deep learning technology such as a large-scale language model. In one example, by using a machine learning model such as deep learning technology, the urgency of the query can be determined using a character string included in the query. As a result of analyzing the query, the server device 10 may display a message 710 in the message area urging the user to contact a predetermined organization if the query is determined to be highly urgent. Alternatively, the server device 10 may urge the user to contact a predetermined organization if the query contains a pre-registered word that can be determined to be highly urgent.

[0067] In one example, the server device 10 may identify a predetermined organization to contact based on the user information of the user who sent the inquiry. For example, the user information shown in FIG. 4(A) includes information identifying the area where the user is likely to be located, such as the user's affiliated organization and address. In such a case, the server device 10 determines a predetermined organization to contact based on the user's area, such as the user's affiliated organization and address, and displays the contact information of the predetermined organization in the message area as shown in FIG. 7(B). This allows the user to understand that the request is urgent and to contact the appropriate organization. In one example, the user ID 401 or other information such as the user name 402 and name 403 may be associated with the user's location.

[0068] <Other embodiments> In the present embodiment, a response is generated based on one piece of handling case data when an inquiry is received. However, a response may be generated based on multiple pieces of handling case data. For example, URLs corresponding to multiple handling cases may be displayed on the screen shown in FIG. 7(A). Alternatively, a response may be generated and displayed based on advice data for multiple handling cases on the screen shown in FIG. 5(C).

[0069] Furthermore, this embodiment and the modified examples can be combined as desired. For example, when the response message shown in FIG. 5A is displayed, a URL for viewing corresponding cases as the source of the response may also be displayed. This allows the user to refer to the link and confirm the basis of the response even if the response text is unclear due to the large-scale language model.

[0070] In addition, in the present embodiment, the server device 10 has been described as providing information about case studies or recommending contact with a specific organization when a response request is received from the client device 20. However, the server device 10 may send a message to the client device 20 requesting input of additional information under predetermined conditions, such as when no case studies highly relevant to the inquiry can be identified or when the number of characters entered is insufficient.

[0071] It is also possible to realize the process of supplying a program that realizes one or more functions of each of the above-mentioned embodiments to a system or device via a network or storage medium, and having one or more processors in the computer of that system or device read and execute the program.

[0072] The invention is not limited to the above-described embodiment, and various modifications and variations are possible within the scope of the gist of the invention. [Explanation of symbols]

[0073] 1: Chat system, 10: Server device, 20: Client device

Claims

1. A server device that provides a chat service for searching for cases of responses to customers, providing means for providing a user interface to the client device; a storage means for storing case studies of responses to customers and past inquiries; a first presentation means for presenting, via the user interface, character strings that are frequently included in the past queries; a reception unit that receives an input of an inquiry about a case example from a customer from the client device via the user interface; an acquisition means for acquiring past case data highly relevant to the inquiry in response to the reception means receiving a response request for case data; a second presentation means for presenting at least a part of the corresponding case data acquired by the acquisition means via the user interface; A server device comprising:

2. The system further comprises a determination unit that determines whether or not the inquiry accepted by the accepting unit recommends contacting a predetermined contact point when a response request for a case study is accepted via the user interface, The server device according to claim 1, characterized in that, when the determination means determines that contacting the specified contact point is recommended, the second presentation means displays a message urging contacting the specified contact point.

3. a second acquisition means for acquiring a query being input to the user interface at a predetermined timing; an update means for updating the first presentation means so as to present character strings that are highly relevant to the query being input and that are acquired by the second acquisition means and that are frequently included in the past queries; 2. The server device according to claim 1, further comprising:

4. The server device according to claim 1, characterized in that, when a selection of a character string presented by the first presentation means is accepted via the user interface, the acquisition means acquires the selected character string and past case data that is highly relevant to the query.

5. the accepting means displays an inquiry currently being input to the user interface; The server device according to claim 1, characterized in that the first presentation means, when accepting a selection of a presented character string via the user interface, adds the selected character string to the query being input.

6. The case data includes consultation data indicating consultations regarding customer responses and advice data indicating advice corresponding thereto; 2. The server device according to claim 1, wherein the acquiring means identifies a degree of association between the consultation data and the inquiry, and acquires the advice data corresponding to the consultation data having the highest degree of association with the inquiry.

7. The server device according to claim 6, characterized in that the acquisition means inputs the query and identifies the corresponding case data that is highly relevant to the query using a machine learning model that outputs information that can identify the corresponding case data that is highly relevant to the query.

8. A control method executed by a server device that has a storage unit that stores case studies of responses to customers and past inquiries, and provides a chat service that searches for case studies of responses to customers, comprising: providing a user interface to the client device; presenting character strings that are frequently included in the past queries via the user interface; receiving an input of an inquiry about a case study from a customer from the client device via the user interface; Upon receiving a response request for a case study, acquiring past case study data highly relevant to the inquiry; presenting at least a portion of the acquired corresponding case data via the user interface; A control method comprising:

9. A computer of a server device that provides a chat service for searching for cases of responses to customers, causing the client device to provide a user interface; It stores case studies of responses to customers and past inquiries, displaying, via the user interface, character strings that are frequently included in the past queries; Accepting an input of an inquiry about a case study from a customer from the client device via the user interface; Upon receiving a response request for a case study, acquire past case study data that is highly relevant to the inquiry; presenting at least a part of the acquired corresponding case data via the user interface; A program characterized by:

Images