Automatic response device and program

Abstract

Reduce the time required to generate responses to inquiries. [Solution] The system determines whether the inquiry matches any of the pre-configured inquiry contents. If the inquiry matches any of the pre-configured inquiry contents, it outputs a first response that represents a pre-configured answer corresponding to the pre-configured inquiry that matches the inquiry. If the inquiry does not match any of the pre-configured inquiry contents, it outputs a second response based on an answer generated by a specific agent using generation AI for the inquiry.

Description

Technical Field

[0001] The present invention relates to a technique for performing automatic responses using Generative Artificial Intelligence.

Background Art

[0002] Techniques for performing automatic responses using Generative Artificial Intelligence are known (see, for example, Non-Patent Document 1, etc.). In such techniques, it is common to generate response contents for inquiry contents using Generative Artificial Intelligence. Also known is a Retrieval Augmented Generation (RAG) technique in which a function called an agent searches for specific data (knowledge information) based on the inquiry content of a user, and generates a response content using at least the search result and Generative Artificial Intelligence.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By generating response contents for inquiry contents using Generative Artificial Intelligence, it is possible to perform appropriate responses to various inquiries. However, there are cases where it takes time to generate response contents using Generative Artificial Intelligence.

[0005] This invention provides a technology for reducing the time required to obtain responses to inquiries. [Means for solving the problem]

[0006] In this invention, it is determined whether the inquiry content matches any of the pre-configured inquiry content. If the inquiry content matches any of the pre-configured inquiry content, a first response content representing a pre-configured response content corresponding to the pre-configured inquiry content that matches the inquiry content is output. If the inquiry content does not match any of the pre-configured inquiry content, a second response content based on a response content generated by a specific agent using generation AI for the inquiry content is output. [Effects of the Invention]

[0007] This reduces the time required to obtain responses to inquiries. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a block diagram illustrating the configuration of an automated response system according to an embodiment. [Figure 2] Figure 2 is a block diagram illustrating the terminal device of the embodiment. [Figure 3] Figure 3 is a block diagram illustrating an automatic response device according to an embodiment. [Figure 4] Figure 4 is a diagram illustrating pre-configured inquiry content and corresponding pre-configured response content. [Figure 5] Figure 5 is a flowchart illustrating the automated response process of the embodiment. [Figure 6] Figure 6 is a diagram illustrating the display screen of an embodiment. [Figure 7] Figure 7 is a diagram illustrating the display screen of the embodiment. [Figure 8] Figure 8 is a block diagram illustrating the hardware configuration of the embodiment. [Modes for carrying out the invention]

[0009] Embodiments of the present invention will be described below with reference to the drawings. [First Embodiment] <Overall Structure> As illustrated in Figure 1, the automated response system 1 of this embodiment includes terminal devices 11-1, ..., 11-M, an automated response device 12, and a generating AI device 13. Hereafter, each of the terminal devices 11-1, ..., 11-M will be referred to as terminal device 11-m. Here, m = 1, ..., M, where M is an integer greater than or equal to 1, for example, an integer greater than or equal to 2. Also, terminal devices 11-1, ..., 11-M will be collectively referred to as terminal device 11. Each terminal device 11-m is a device used by user 1011-m. Hereafter, users 1011-1, ..., 1011-M will be collectively referred to as user 1011. The automated response device 12 is configured to communicate with terminal devices 11 via a network such as the Internet. The automated response device 12 is also configured to be able to call the generating AI device 13 and use the generating AI device 13. User 1011-m makes a query to the automated response device 12 using terminal device 11-m, and the automated response device 12 obtains an answer to the query using the generating AI device 13 and responds to terminal device 11-m. The automated response system 1 illustrated in Figure 1 has one automated response device 12 and one generating AI device 13. However, the automated response system 1 may have multiple automated response devices 12 or multiple generating AI devices 13.

[0010] <Terminal device 11-m> As illustrated in Figure 2, the terminal device 11-m in this embodiment includes a user interface 111-m, a communication unit 112-m, a control unit 116-m, and a memory 117-m. The terminal device 11-m performs each process under the control of the control unit 116-m. Information input to the terminal device 11-m and information obtained by each part of the terminal device 11-m are stored in the memory 117-m and read out and used as needed. Examples of the terminal device 11-m include smartphones, smartwatches, tablet terminals, personal computers, etc., with a predetermined program loaded. The user interface 111-m is, for example, a touchscreen, or a combination of a display, mouse, keyboard, touch panel, speaker, microphone, etc.

[0011] <Automatic response device 12> As illustrated in Figure 3, the automatic response device 12 of this embodiment includes a storage unit 120, a communication unit 121, a calling unit 123, a dispatcher 124, agents 125-1, ..., 125-N, a control unit 126, and a memory 127. N is an integer greater than or equal to 1, for example, N is an integer greater than or equal to 2. Each of the agents 125-1, ..., 125-N is denoted as agent 125-n, where n = 1, ..., N. Agents 125-1, ..., 125-N are collectively referred to as agent 125. The set whose elements are agents 125-1, ..., 125-N is called the set of agents. The dispatcher 124 includes a determination unit 1241, response units 1242, 1244, and a selection unit 1243. Agent 125-n includes a search unit 1251-n and a generation unit 1252-n. The search units 1251-1, ..., 1251-N are collectively referred to as the search unit 1251, and the generation units 1252-1, ..., 1252-N are collectively referred to as the generation unit 1252. The automatic response device 12 executes each process under the control of the control unit 126. Information input to the automatic response device 12, and information obtained by each part of the automatic response device 12, are stored in the memory 127 and read out and used as needed. An example of the automatic response device 12 is a server device or a device configured on the cloud with a predetermined program loaded.

[0012] <Generation AI device 13> The generative AI device 13 is a device that implements generative AI based on a machine learning model and has the function of generating and returning a response to given information. For example, the generative AI generates text based on a large language model (LLM). In addition, the generative AI may also generate images based on an image generation model, generate videos based on a video generation model, generate audio based on a speech generation model, or generate music based on a music generation model. An example of the generative AI device 13 is a server device with a predetermined program loaded, or a device configured on the cloud.

[0013] <Pre-configuration> This section describes the prerequisites for this form of automated response processing. The following prerequisites are described here. (1) Pre-configuration of Agent 125 (2) Pre-configuration of dispatcher 124

[0014] (1) Pre-configuration of Agent 125 Agent 125 is a processing unit that retrieves pre-configured data (knowledge information) based on the query content, obtains search results, and uses at least the search results and a generating AI to obtain an answer corresponding to the query content. For example, Agent 125 may provide the generating AI with the search results or information based thereon to obtain an answer from the generating AI, or it may provide the generating AI with the query content and the search results or information based thereon to obtain an answer from the generating AI. The data that each agent 125-n can search is pre-configured, which determines the properties of each agent 125-n. That is, the data that each agent 125-n can search functions as knowledge for that agent 125-n, and the answer obtained using at least the search results and the generating AI is based on the knowledge of that agent 125-n. In other words, setting the data that each agent 125-n can search is equivalent to setting the knowledge for each agent 125-n. Preferably, the data that the same agent 125 can search has commonalities, and the data that different agents 125 can search has differences. For example, the same agent 125 may be able to search for data in the same field, while different agents 125 may be able to search for data in different fields. Specifically, for example, agent 125-1 may be able to search only for data in the IT field, agent 125-2 may be able to search only for data in the legal field, and agent 125-3 may be able to search only for data in the economic field. For example, the same agent 125 may be able to search for data on the same product, while different agents 125 may be able to search for data on different products. Specifically, for example, agent 125-1 may be able to search only for data on telephone and fax machines, agent 125-2 may be able to search only for data on smartphones, and agent 125-3 may be able to search only for data on automobiles. For example, the same agent 125 may be able to search for data on the same model, while different agents 125 may be able to search for data on different models.For example, agent 125-1 may be able to search only for data related to model A of the telephone & fax machine, agent 125-2 may be able to search only for data related to model B of the telephone & fax machine, and agent 125-3 may be able to search only for data related to model C of the telephone & fax machine. The data may be document data, text data, table-form data, spreadsheet data, or table data that can be separated by columns and rows, text data that can be separated by chapters, paragraphs, sentences, etc., XML (Extensible Markup Language) data with tags, CSV (comma separated values) file data that can be separated by entries, program code data separated by comment-outs, graphic data, video data, or audio data. The data may be non-public information or public information. In this embodiment, at least some of the data includes FAQ data representing a set of pre-set query contents and pre-set answer contents corresponding to the pre-set query contents (hereinafter referred to as "the set of pre-set query contents and answer contents"). The set of pre-set query contents and answer contents is, for example, what is called "Frequently Asked Questions (FAQ)". FIG. 4 illustrates a set of pre-set query contents and answer contents. In FIG. 4, a set of a pre-set query content "What kind of service is "〇〇〇"?" and a pre-set answer content ""〇〇〇" is a beneficial plan in which six optional services and call charges are set", and a set of a pre-set query content "Please tell me the conditions for free use of "△△△"." and a pre-set answer content "To make "△△△" free of charge, a contract for "□□□" is required." are illustrated. In addition, the data may represent, for example, the business knowledge of a specific organization, know-how, expertise, or a product manual.

[0015] In the pre-configuration of agent 125, data searchable by agent 125-n (n=1,…,N) is stored in the storage unit 120 of the automatic response device 12. Hereafter, data searchable by agent 125-n will be referred to as data 1212-n, and data 1212-1,…,1212-N will be collectively referred to as data 1212. At least a portion of data 1212-1,…,1212-N stored in the storage unit 120 includes FAQ data. That is, data 1212 searchable by at least one of one or more agents 125 includes information representing the pre-configured inquiry content and information representing the pre-configured response content. In addition to data 1212-n, other registration information 1213-n related to data 1212-n and / or agent 125-n may also be stored in the storage unit 120. For example, registration information 1213-n may include summary information corresponding to data that agent 125-n can search, or it may include information representing search conditions (search parameters) when searching for data. The summary information may be, for example, summary information for agent 125-n, or summary information for data 1212-n that agent 125 can search. For example, the summary information is information that is searched in order to select an agent suitable for the query content.

[0016] (2) Pre-configuration of dispatcher 124 The dispatcher 124 of this embodiment is a processing unit that plays a role in providing a comprehensive guide for automatic response. The dispatcher 124 has a function of selecting a specific agent 125 that enables the agent 125 to obtain the response content (generated response content) for the inquiry content by using the input information corresponding to the inquiry content and searching for the information corresponding to the data 1212 that can be searched by the agent 125. Further, the dispatcher 124 has a function of directly accessing at least the FAQ data among the data 1212 stored in the storage unit 120. When the input inquiry content matches any of the FAQ data, the dispatcher 124 obtains the response content represented by the FAQ data that matches the input inquiry content instead of causing the agent 125 to obtain the response content. Thereby, the acquisition time of the response content for the inquiry content can be shortened. In the presetting of the dispatcher 124, the determination criteria when the dispatcher 124 searches for information are set. For example, the determination criteria when the dispatcher 124 searches for FAQ data may be set, or the determination criteria when searching for the agent 125 may be set. The determination criteria for searching may include, for example, the tolerance (tolerance) in vector search, or may include the conditions for text search such as exact match search and partial match search.

[0017] <Automatic response processing> Next, the automatic response processing of this embodiment will be described using FIG. 5. User 1011-m inputs the inquiry into the user interface 111-m of terminal device 11-m (Figure 2). The inquiry can be entered as text or voice. For example, a display screen 1100, as illustrated in Figures 6 and 7, is displayed on the user interface 111-m, and user 1011-m inputs the inquiry into the input field 1110 of the display screen 1100. For example, in the example in Figure 6, the inquiry is "What is...?" and in the example in Figure 7, the inquiry is "What happens if... and... are added to...?" The information representing the entered inquiry is sent to the communication unit 112-m. The communication unit 112-m transmits the information representing the inquiry to the automatic response device 12 via the network. The communication unit 121 of the automatic response device 12 (Figure 3) receives the information representing the inquiry and sends it to the dispatcher 124. The determination unit 1241 of the dispatcher 124 receives information representing the content of the inquiry (step S101).

[0018] The determination unit 1241 of the dispatcher 124 accesses the FAQ data contained in the data 1212 stored in the storage unit 120 and determines whether the inquiry sent from the communication unit 121 matches any of the pre-configured inquiry contents (inquiry contents of the FAQ data). For example, the determination unit 1241 of the dispatcher 124 determines whether the inquiry matches any of the pre-configured inquiry contents based on the pre-configured determination criteria as described above (step S102).

[0019] In step S102, if it is determined that the inquiry content matches any of the pre-configured inquiry content, the determination unit 1241 extracts the pre-configured response content corresponding to the pre-configured inquiry content that matches the inquiry content from the storage unit 120 and sends the response content (first response content) representing it to the response unit 1242 (first response unit). The response unit 1242 outputs the response content (first response content) to user 1011-m that represents the response content to the communication unit 121. This response content to user 1011-m may include additional information indicating that it is a pre-configured response content. The communication unit 121 transmits the information representing the response content to user 1011-m to the terminal device 11-m via the network. The communication unit 112-m of the terminal device 11-m (Figure 2) receives the information representing the response content to user 1011-m and sends it to the user interface 111-m. The user interface 111-m outputs the response content to user 1011-m. For example, the user interface 111-m may display the answer to user 1011-m or output it as audio. For example, in the example in Figure 6, the display screen 1100 is displayed on the user interface 111-m, and the answer (first answer) is displayed in the answer field 1120. In this example, the answer displayed in the answer field 1120 is "'XXX' is a great value plan that includes six optional services and call charges." Furthermore, if the answer to user 1011-m includes additional information indicating that it is a pre-configured answer, the user interface 111-m may output a statement indicating that it is a pre-configured answer in addition to the answer. In the example in Figure 6, the answer field 1120 displays the mark "FAQ" 1121 indicating that it is a pre-configured answer. Then, proceed to step S107 (step S103).

[0020] On the other hand, if in step S102 it is determined that the query content does not match any of the pre-configured query content, the selection unit 1243 of the dispatcher 124 selects a specific agent 125-s that is suitable for the query content from among agents 125-1, ..., 125-N (s∈{1, ..., N}). A specific agent 125-s belongs to a set of one or more agents 125-1, ..., 125-N. If N=1, the selection unit 1243 selects agent 125-1. If N≧2, the selection unit 1243 selects a specific agent 125-s that is suitable for the query content from among multiple agents 125-1, ..., 125-N based on the input information corresponding to the query content. That is, the selection unit 1243 uses the input information corresponding to the query content to search for information corresponding to the data 1212 that multiple agents can each search, and selects a specific agent 125-s. This search may be a vector search, a keyword search, another type of search, or a hybrid search combining multiple search methods. The number of specific agents 125-s selected may be singular or plural. The number of specific agents 125-s C may be defined, or an upper limit Cmax may be defined for the number of specific agents 125-s, or neither may be defined. Note that Cmax is an integer greater than or equal to 1. The following is an example of how to select specific agents 125-s when N≧2.

[0021] <Example 1 of how to select a specific agent 125-s> (1-1) The selection unit 1243 uses a search query representing the query content (input information corresponding to the query content) to search for information corresponding to the data 1212-1, ..., 1212-N stored in the storage unit 120. The information corresponding to data 1212-1, ..., 1212-N may be, for example, data 1212-1, ..., 1212-N itself, function values ​​(excluding constant function values) derived from each of the data 1212-1, ..., 1212-N, registration information 1213-1, ..., 1213-N, a part of each of the registration information 1213-1, ..., 1213-N (for example, summary information), function values ​​(excluding constant function values) derived from each of the registration information 1213-1, ..., 1213-N, or function values ​​(excluding constant function values) derived from a part of each of the registration information 1213-1, ..., 1213-N (for example, summary information). The function value may be a scalar value or a vector value. For example, the selection unit 1243 uses a vector representing the query content as the search query and performs a vector search for vectors representing the summary information of each of the registration information items 1213-1, ..., 1213-N. The criteria for the search may be fixed or pre-set as described above. (1-2) If the number of hits in step (1-1) exceeds the upper limit Cmax, proceed to step (1-3). On the other hand, if the number of hits in step (1-1) is less than or equal to the upper limit Cmax, proceed to step (1-5). (1-3) The selection unit 1243 communicates with the terminal device 11-m (for example, via chat) in order to narrow down the number of hits. Specifically, the selection unit 1243 sends information representing questions to the communication unit 121 to narrow down the number of hits, and the communication unit 121 transmits the information representing the questions to the terminal device 11-m via the network. The communication unit 112-m of the terminal device 11-m (Figure 2) receives the information representing the questions and sends it to the user interface 111-m, and the user interface 111-m outputs the questions. The user 111-m inputs the answers to the questions into the user interface 111-m. The answers are sent to the communication unit 112-m and transmitted to the automatic response device 12 via the network. The communication unit 121 of the automatic response device 12 (Figure 3) receives the answers and sends them to the selection unit 1243. The selection unit 1243 uses a search query that represents information corresponding to the response content and the inquiry content (input information corresponding to the inquiry content) to search for information corresponding to data 1212-1, ..., 1212-N stored in the storage unit 120. This search method is the same as in procedure (1-1). (1-4) If the number of hits in step (1-3) exceeds the upper limit Cmax, return to step (1-3). On the other hand, if the number of hits in step (1-3) is less than or equal to the upper limit Cmax, proceed to the next step (1-5). (1-5) The selection unit 1243 designates the agent 125 corresponding to the information found in the search as a specific agent 125-s.

[0022] <Example 2 of how to select a specific agent 125-s> (2-1) The selection unit 1243 uses a search query representing the inquiry content (input information corresponding to the inquiry content) to search for information corresponding to the data 1212-1, ..., 1212-N stored in the storage unit 120. This search method is the same as in procedure (1-1). (2-2) The selection unit 1243 outputs up to Cmax agents 125 that correspond to the information hit in the search, selected in order of the degree of matching (e.g., relevance score, similarity, etc.) as a specific agent 125-s. For example, when the selection unit 1243 performs a vector search for vectors corresponding to data 1212-1, ..., 1212-N (e.g., vectors representing summary information) using a vector representing the query content as the search query, it selects up to Cmax agents 125 that correspond to the vectors selected in order of the degree of similarity to the search query as a specific agent 125-s.

[0023] <Example 3 of how to select a specific agent 125-s> Information representing the specific agent 125-s selected in Example 1 or Example 2 of the method for selecting a specific agent 125-s may be sent to the terminal device 11-m, and the agents 125-s may be narrowed down based on the selection by the user 1011-m. That is, the selection unit 1243 sends information representing agent 125-s to the communication unit 121, and the communication unit 121 transmits information representing agent 125-s to the terminal device 11-m via the network. The communication unit 112-m of the terminal device 11-m (Figure 2) receives the information representing agent 125-s and sends it to the user interface 111-m, and the user interface 111-m outputs information representing agent 125-s. The user 1011-m selects some or all of the outputted agents 125-s and inputs information representing the selection into the user interface 111-m. The information representing the selection is sent to the communication unit 112-m and transmitted to the automatic response device 12 via the network. The communication unit 121 of the automatic response device 12 (Figure 3) receives information representing the selection and sends it to the selection unit 1243. Based on the information representing the selection, the selection unit 1243 sets the final specific agent 125-s.

[0024] The selection unit 1243 (Figure 3) sends information representing the specific agent 125-s described above to the response unit 1244 (step S104).

[0025] The response unit 1244 (second response unit) outputs a response (second response) based on the generated response obtained by a specific agent 125-s (s∈{1,…,N}) using a generation AI, for the query content. That is, each agent 125 has the function of obtaining a generated response using at least the search results obtained by searching the data 1212 that it can search based on the query content, and the generation AI. The response unit 1244 controls a specific agent 125-s, obtains a generated response for the query content, and outputs a response based on that generated response. For example, the response unit 1244 sends information representing the query content to each of the specific agents 125-s mentioned above and makes a query to each agent 125-s. The search unit 1251-s of agent 125-s uses the information representing the query content to search the data 1212-s stored in the storage unit 120 and extracts the search results, which are the hit information. This search may be a vector search, a keyword search, another type of search, or a hybrid search combining multiple search methods. The search conditions may be fixed or represented in the registration information 1213-s. It is desirable that the search results be semantically coherent units that can be divided according to their structure. Examples of such search results include tabular data, spreadsheet data, column and row data of table data, chapter, paragraph and sentence data of document data, tag-separated data of XML data, entry data of CSV file data, and code data separated by comments in program code data. The search results may contain a single semantically coherent unit or multiple semantically coherent units. The search results are sent to the generation unit 1252-s of agent 125-s. The generation unit 1252-s calls the generation AI device 13 via the calling unit 123 (API (Application Programming Interface)) and uses the information representing the query content, the data based on the search results, and the generation AI of the generation AI device 13 to obtain a generated response content corresponding to the query content.In other words, the generation unit 1252-s uses at least the search results and the generation AI to obtain a generated response corresponding to the query. For example, the generation unit 1252-s sends data (e.g., payload) based on information representing the query and the search results to the generation AI device 13 via the call unit 123. The generation AI device 13 generates a generated response for the data based on information representing the query and the search results, and returns the generated response (e.g., payload) to the generation unit 1252-s via the call unit 123. If the search results can be divided into semantically coherent units by their structure, preferably, the generation unit 1252-s uses at least the search results divided into semantically coherent units by their structure (e.g., rows of tabular data, etc.) and the generation AI to obtain a generated response corresponding to the query. The generation unit 1252-s sends the obtained generated response to the response unit 1244 of the dispatcher 124. The dispatcher 124's response unit 1244 obtains and outputs a response to user 1011-m (second response) based on the generated response content sent from each agent 125-s. For example, the response unit 1244 may use a summary of the generated response content sent from agent 125-s as the response to user 1011-m, or use any of the generated response content sent from agent 125-s as the response to user 1011-m, or use a new response content generated from the generated response content sent from agent 125-s as the response to user 1011-m. For example, the response unit 1244 may call the generation AI device 13 via the call unit 123 and obtain a response to user 1011-m using the generated response content sent from agent 125-s and the generation AI of the generation AI device 13. This response to user 1011-m may include additional information indicating that it is based on content generated by the generation AI. Information representing the response to user 1011-m is sent to the communication unit 121. The communication unit 121 transmits the information representing the response to user 1011-m to the terminal device 11-m via the network. The communication unit 112-m of the terminal device 11-m (Figure 2) receives the information representing the response to user 1011-m and sends it to the user interface 111-m.The user interface 111-m outputs the response content to user 1011-m. For example, the user interface 111-m may display the response content to user 1011-m or output it as audio. For example, the display screen 1100 illustrated in Figure 7 is displayed on the user interface 111-m, and the response content (second response content) is displayed in the answer field 1120 of the display screen 1100. In the example in Figure 7, the response content displayed in the answer field 1120 is "Regarding..., if... and... are added together...". Furthermore, if the response content to user 1011-m includes additional information indicating that it is based on content generated by the generating AI, the user interface 111-m may output in addition to the response content that it is based on content generated by the generating AI. In the example in Figure 7, the mark "AI" 1122 indicating that it is based on content generated by the generating AI is displayed in the answer field 1120. Then proceed to step S107 (step S105).

[0026] In step S107, the determination unit 1241 (Figure 3) determines whether or not it has received a new inquiry from user 1011-m. For example, if the determination unit 1241 receives information representing a new inquiry from terminal device 11-m after the response content to user 1011-m has been output from the response unit 1242 or response unit 1244 (step S103 or S105), it determines that it has received a new inquiry from user 1011-m. For example, if the display screen 1100 illustrated in Figure 6 or Figure 7 is displayed on the user interface 111-m of terminal device 11-m (Figure 2) (step S103 or S105), and the response content is displayed in the answer field 1120, user 1011-m can use the user interface 111-m to input a new inquiry into the input field 1130. In the examples in Figures 6 and 7, user 1011-m may enter the inquiry content as text in the input field 1130, select the inquiry content from the pre-configured inquiry content (e.g., frequently asked questions) displayed in the tag unit 1131, or add the entered text content to the inquiry content selected from the tag unit 1131. The pre-configured inquiry content displayed in the tag unit 1131 may or may not be the same as the FAQ data included in data 1212 (Figure 3). The information representing the entered inquiry content is sent to the communication unit 112-m of the terminal device 11-m (Figure 2). The communication unit 112-m transmits the information representing the inquiry content to the automatic response device 12 via the network. The communication unit 121 of the automatic response device 12 (Figure 3) receives the information representing the inquiry content and sends it to the dispatcher 124. The determination unit 1241 of the dispatcher 124 accepts the information representing the inquiry content. For example, in this case, the determination unit 1241 determines that it has received a new inquiry from user 1011-m. If it is determined that it has received a new inquiry from user 1011-m, the process returns to step S102.On the other hand, for example, if the determination unit 1241 does not receive any information representing new inquiry content from the terminal device 11-m even after a predetermined time has elapsed since the response content to user 1011-m was output from the response unit 1242 or 1244 (step S103 or S105), the determination unit 1241 determines that no new inquiry content was received from user 1011-m. In this case, the automatic response process ends.

[0027] <Features of this form> In this configuration, the determination unit 1241 of the dispatcher 124 of the automated response device 12 (Figure 3) determines whether the inquiry content matches any of the "pre-configured inquiry content". If the inquiry content matches any of the pre-configured inquiry content, the response unit 1242 (first response unit) outputs a response content (first response content) that represents the "pre-configured response content" corresponding to the "pre-configured inquiry content" that matches the inquiry content. This shortens the time required to obtain the response content for an inquiry. On the other hand, if the inquiry content does not match any of the "pre-configured inquiry content", the response unit 1244 (second response unit) outputs a response content (second response content) based on the generated response content obtained by a specific agent 125-s using a generation AI for the inquiry content. This enables appropriate responses to a variety of inquiries. Thus, in this configuration, appropriate responses can be provided to a variety of inquiries, and the time required to obtain the response content for an inquiry can be shortened.

[0028] In this configuration, a specific agent 125-s belongs to a set of one or more agents 125-1, ..., 125-N. Each of the one or more agents 125 has the function of obtaining generated response content by searching for searchable data 1212 based on the query content and using at least the search results and generating AI. The searchable data 1212 (knowledge information) of at least one of the one or more agents 125 includes information representing pre-configured query content and information representing pre-configured response content (FAQ data). That is, the response content based on the generated response content obtained by the agent (second response content) is based on data 1212 (knowledge information), and the response content representing the pre-configured response content (first response content) is based on the FAQ data included in data 1212 (knowledge information). Therefore, the consistency between the response content representing the pre-configured response content (first response content) and the response content based on the generated response content obtained by the agent (second response content) is improved, and the overall uniformity of the response content is improved.

[0029] [Second Embodiment] The second embodiment is a modification of the first embodiment. The completeness of the FAQ data included in data 1212 varies. The amount of answers in the FAQ data included in each data 1212-n may be large or small, and the quality of the answers may be high or low. Generally, the more complete the answers in the FAQ data, the more appropriate they are as answers to user 1011. Taking this into consideration, the dispatcher in this embodiment determines whether the inquiry matches any of the pre-set inquiry contents based on a judgment criterion that represents the completeness of the pre-set answer contents (FAQ data). Here, the frequency at which the index determines that the pre-set answer contents with a first value match the inquiry is higher than the frequency at which the index determines that the pre-set answer contents with a second value match the inquiry. The completeness represented by the first value is higher than the completeness represented by the second value. This improves the balance between the quality of the answer contents and the shortness of the response time, and the overall quality of the automated response is improved. In the following, the differences from the first embodiment will be explained in detail, and the explanation of matters already explained will be simplified by reusing the same reference numbers.

[0030] <Overall Structure> As illustrated in Figure 1, the automated response system 2 of this embodiment includes terminal devices 11-1, ..., 11-M, an automated response device 22, and a generating AI device 13. The automated response device 22 is configured to communicate with terminal devices 11 via a network such as the Internet. The automated response device 22 is also configured to call and utilize the generating AI device 13. A user 1011-m makes a query to the automated response device 22 using terminal device 11-m, and the automated response device 22 obtains an answer to the query using the generating AI device 13 and responds to terminal device 11-m. Note that the automated response system 2 illustrated in Figure 1 has one automated response device 22 and one generating AI device 13. However, the automated response system 2 may have multiple automated response devices 22 or multiple generating AI devices 13.

[0031] <Automatic response device 22> As illustrated in Figure 3, the automatic response device 22 in this embodiment includes a storage unit 120, a communication unit 121, a calling unit 123, a dispatcher 224, agents 125-1, ..., 125-N, a control unit 126, and a memory 127. The dispatcher 224 includes a determination unit 2241, response units 1242, 1244, and a selection unit 1243. The automatic response device 22 executes each process under the control of the control unit 126. Information input to the automatic response device 22 and information obtained by each part of the automatic response device 22 are stored in the memory 127 and read out and used as needed. An example of the automatic response device 22 is a server device or a device configured on the cloud with a predetermined program loaded.

[0032] <Pre-configuration> The differences from the pre-configuration of the first embodiment are as follows:

[0033] (1) Pre-configuration of Agent 125 In this configuration, at least the storage unit 120 stores the registration information 1213-n. However, each piece of registration information 1213-n in this configuration includes at least an index i(n) representing the completeness of the pre-configured response content included in each piece of data 1212-n. The index i(n) representing completeness may represent, for example, the quantity of the pre-configured response content included in the data 1212-n, or its quality, or a combination thereof. For example, a larger quantity of response content may indicate a higher completeness, or a higher quality response content may indicate a higher completeness. The quantity of pre-configured response content may represent, for example, the number of pre-configured response items (e.g., the number of pairs of pre-configured inquiry content and response content), the length (e.g., the total number of characters in the response content), the number of links included in the response content (e.g., the total number of links included in the response content), or at least a combination of these. The pre-configured quality of the response content may, for example, represent the satisfaction level of the response content, the percentage of inquiries completed based on the response content, or a combination of these. Otherwise, it is the same as the pre-configuration of agent 125 in the first embodiment, except that the automatic response device 12 is replaced by the automatic response device 22.

[0034] (2) Pre-configuration of dispatcher 224 In this embodiment, the dispatcher 224 determines whether the inquiry content matches any of the pre-configured inquiry contents based on a judgment criterion that uses an index i(n) representing the completeness of the pre-configured response content. This judgment results in a higher frequency of determining that a pre-configured response content (FAQ data included in data 1212-n) with index i(n) α1 (first value) matches the inquiry content than a higher frequency of determining that a pre-configured response content (FAQ data included in data 1212-n) with index i(n) α2 (second value) matches the inquiry content. Note that the completeness represented by α1 (first value) is higher than the completeness represented by α2 (second value). For example, among the FAQ data, the more complete the data, the higher the frequency of determining that it matches the inquiry content. In the pre-configuration of the dispatcher 224 in this embodiment, at least this judgment criterion is set. Otherwise, the pre-configuration is the same as that of the dispatcher 124 in the first embodiment, except that the automatic response device 12 is replaced by the automatic response device 22.

[0035] The following are examples of evaluation criteria based on the index i(n) representing the level of completeness. <Example of judgment criteria> For example, the index i(n) representing the degree of fulfillment can be defined as follows: i(n) = (Update frequency score × W1) + (Average feedback score × W2) + (Number of items score × W3) In this example, a larger index i(n) indicates a higher level of completeness of the FAQ data included in the dataset 1212-n. The update frequency score represents the update frequency of the FAQ data included in data 1212-n, for example. A higher update frequency score indicates a higher update frequency of the FAQ data. For example, the update frequency score can be determined as follows: • If recently updated: +10 points • A few months ago: +5 points • More than a year ago: +0 points If data 1212-n contains multiple FAQ data, for example, the sum of the update frequency scores of each of those multiple FAQ data may be used, or the average value may be used. The average feedback score is, for example, the average feedback score for the answers represented by the FAQ data included in data 1212-n. A higher feedback score indicates a better evaluation of the answer. For example, the average feedback score can be determined as follows: ·4.5-5.0: +10 points ·3.5-4.4: +8 points ·2.5-3.4: +5 points ·1.5-2.4: +2 points ·1.0-1.4: +0 points The count score is based on the number of FAQ data entries included in data set 1212-n, for example. A higher FAQ count score indicates a larger number of FAQ data entries in data set 1212-n. For example, the count score can be determined as follows: ·30 or more: +10 points • 10 to less than 30 items: +5 points • Less than 10 items: +0 points W1, W2, and W3 are weights representing the importance of each of the scores mentioned above. For example, the weights W1, W2, and W3 are defined as follows: • W1 (Importance of update frequency score): 0.3 • W2 (Importance of average feedback score): 0.6 • W3 (Importance of the number of cases score): 0.1 For example, if the update frequency score is +10 points, the average feedback score is +8 points, the number of responses score is +10 points, and W1=0.3, W2=0.6, and W3=0.1, the index i(n) representing the level of completeness is calculated as follows. i(n)=(10×0.3)+(8×0.6)+(10×0.1)=3+4.8+1=8.8 The dispatcher 224 in this configuration determines whether the inquiry content matches any of the pre-configured inquiry contents based on a judgment criterion such as the index i(n). For example, the dispatcher 224 determines that the inquiry content matches the pre-configured inquiry content if the similarity between the inquiry content and the pre-configured inquiry content exceeds a threshold based on index i(n), and determines that it does not match otherwise. However, this threshold is set to decrease as the index i(n) increases. As a result, the frequency at which a pre-configured response content with index i(n) α1 is determined to match the inquiry content is higher than the frequency at which a pre-configured response content with index i(n) α2 is determined to match the inquiry content. The level of completeness represented by α1 is higher than the level of completeness represented by α2.

[0036] <Automated response processing> The automatic response processing in this embodiment replaces step S102 of the automatic response processing in the first embodiment with the following step S202 (Figure 5). Otherwise, it is the same as the first embodiment, except that the automatic response device 12 and dispatcher 124 are replaced by automatic response device 22 and dispatcher 224, respectively.

[0037] Step S202: The determination unit 2241 of the dispatcher 224 (Figure 3) accesses the FAQ data contained in the registration information 1213 and data 1212 stored in the storage unit 120, and determines whether the inquiry sent from the communication unit 121 matches any of the pre-set inquiry contents (inquiry contents in the FAQ data) based on the determination criteria based on the index i(n) representing the completeness of the registration information 1213-n. As a result of this determination, the frequency in which the pre-set answer content with index i(n) α1 (first value) is determined to match the inquiry content is higher than the frequency in which the pre-set answer content with index i(n) α2 (second value) is determined to match the inquiry content. The completeness represented by α1 (first value) is higher than the completeness represented by α2 (second value). The determination criteria based on the index i(n) representing the completeness are as described above. For example, in the case of the aforementioned example of the judgment criteria, the judgment unit 2241 of the dispatcher 224 determines that the inquiry content matches the pre-configured inquiry content if the similarity between the inquiry content and the pre-configured inquiry content exceeds a threshold based on the index i(n), and determines that it does not match otherwise (step S202).

[0038] <Features of this form> In this configuration, the determination unit 2241 of the dispatcher 224 of the automatic response device 22 (Figure 3) determined whether the inquiry sent from the communication unit 121 matched any of the "pre-configured inquiry contents" based on a determination criterion that uses an index i(n) representing the completeness of the "pre-configured response contents". In this configuration, the frequency at which the pre-configured response contents with index i(n) α1 (first value) are determined to match the inquiry contents is higher than the frequency at which the pre-configured response contents with index i(n) α2 (second value) are determined to match the inquiry contents. Note that the completeness represented by α1 (first value) is higher than the completeness represented by α2 (second value). By using a determination criterion based on the completeness of the "pre-configured response contents" in this way, it becomes possible to appropriately determine whether to output the response contents representing the "pre-configured response contents" (first response contents) or the response contents based on the generated response contents obtained by a specific agent 125-s (second response contents). As a result, the balance between the quality of the response and the short response time improves, and the overall quality of automated responses improves.

[0039] [Third Embodiment] For the same user 1011, before a response based on a generated response is output for a given inquiry, a response (third response) representing a pre-configured response for a past inquiry (second inquiry) may be output (Figure 5). In such cases, if the latter response (third response) output earlier can be used when acquiring the former generated response, the consistency of the responses in the series of automated response processing can be improved. From this perspective, the agent in this embodiment acquires the generated response for the former inquiry using at least the latter response (third response), search results, and generation AI. The latter response (third response) represents a pre-configured response (FAQ data response) corresponding to a pre-configured inquiry that matches an inquiry earlier than the former inquiry (second inquiry). Below, we will mainly explain the differences from the first embodiment, and for matters already explained, we will simplify the explanation by reusing the same reference numbers.

[0040] <Overall Structure> As illustrated in Figure 1, the automated response system 3 in this embodiment includes terminal devices 11-1, ..., 11-M, an automated response device 32, and a generating AI device 13. The automated response device 32 is configured to communicate with terminal devices 11 via a network such as the Internet. The automated response device 32 is also configured to call and utilize the generating AI device 13. A user 1011-m makes a query to the automated response device 32 using terminal device 11-m, and the automated response device 32 obtains an answer to the query using the generating AI device 13 and responds to terminal device 11-m. Note that the automated response system 3 illustrated in Figure 1 has one automated response device 32 and one generating AI device 13. However, the automated response system 3 may have multiple automated response devices 32 or multiple generating AI devices 13.

[0041] <Automatic response device 32> As illustrated in Figure 3, the automatic response device 32 in this embodiment includes a storage unit 120, a communication unit 121, a calling unit 123, a dispatcher 324, agents 125-1, ..., 125-N, a control unit 126, and a memory 127. The dispatcher 324 includes a determination unit 1241, response units 3242, 3244, and a selection unit 1243. The automatic response device 32 executes each process under the control of the control unit 126. Information input to the automatic response device 32 and information obtained by each part of the automatic response device 32 are stored in the memory 127 and read out and used as needed. An example of the automatic response device 32 is a server device or a device configured on the cloud with a predetermined program loaded.

[0042] <Pre-configuration> The pre-configuration is the same as in the first embodiment, except that the automatic response device 12 is replaced by the automatic response device 32.

[0043] <Automated response processing> The automatic response processing in this embodiment is the same as the first embodiment except that step S103 of the automatic response processing in the first embodiment is replaced with the following step S303, and step S105 is replaced with the following step S305.

[0044] Step S303: In step S303, the response unit 3242 (first response unit) performs the processing of step S103 described in the first embodiment, instead of the response unit 1242, and further stores the response content (third response content) for user 1011-m obtained in the processing of step S103 in the memory 127 (step S303).

[0045] Step S305: In step S305, instead of the response unit 1244, the response unit 3244 (second response unit) outputs a response (second response) based on the generated response obtained by a specific agent 125-s (s∈{1,…,N}) that uses a generation AI for the query content. However, each agent 125 in this configuration has the function of obtaining the generated response using the search results obtained by searching the data 1212 that it can search based on the query content, the generation AI, and at least past response content (third response) extracted from memory 127. It is desirable that the query content corresponding to the generated response and the past query content (second query content) corresponding to the past response content (third response content) used to obtain the generated response are transmitted from the same terminal device 11. In other words, it is desirable that the query content corresponding to the obtained generated response and the past query content (second query content) are generated by the same user 1011. More preferably, the query content corresponding to the acquired generated response content and the past query content (second query content) are query content received by the dispatcher 324 from the time the dispatcher 324 receives the query content in step S101 until the dispatcher 324 determines in step S107 that it has not received any information representing new query content. In other words, it is desirable that these query contents are received by the dispatcher 324 in a series of automated response processes. However, this does not limit the present invention. For example, the query content corresponding to the acquired generated response content and the past query content (second query content) may be generated by a user 1011 with the same attributes (i.e., they may be transmitted from a terminal device 11 of a user 1011 with the same attributes). Examples of attributes include nationality, occupation, field of expertise, age, gender, length of use, etc. For example, the response unit 3244 sends the past response content (third response content) and information representing the query content to each of the specific agents 125-s described above and makes a query to each agent 125-s.The search unit 1251-s of agent 125-s uses information representing the query content to search the data 1212-s stored in the memory unit 120 and extracts the search results, which are the information that hits the search. This search is performed as described in the first embodiment, for example. The search results are sent to the generation unit 1252-s of agent 125-s. The generation unit 1252-s calls the generation AI device 13 via the call unit 123 and uses the information representing the query content, past answer content (third answer content), data based on the search results, and the generation AI of the generation AI device 13 to obtain generated answer content corresponding to the query content. That is, the generation unit 1252-s obtains generated answer content corresponding to the query content using at least past answer content (third answer content), search results, and the generation AI. For example, the generation unit 1252-s sends the information representing the query content, data based on past answer content (third answer content) and search results to the generation AI device 13 via the call unit 123. The generation AI device 13 generates a generated response based on data representing the inquiry content, past response content (third response content), and search results, and returns it to the generation unit 1252-s via the calling unit 123. The generation unit 1252-s sends the obtained generated response content to the response unit 3244 of the dispatcher 324. The processing in step S305 from here on is the same as the processing in step S105 described in the first embodiment, except that the dispatcher 124 is replaced by the dispatcher 324 and the response unit 1244 is replaced by the response unit 3244 (step S305).

[0046] <Features of this form> In this configuration, a specific agent 125-s (s∈{1,…,N}) obtains generated response content using at least the following: a response content (third response content) representing a "pre-configured response content" corresponding to a "pre-configured response content" that matches a past query content (second query content) that is more similar to the current query content; and a generating AI. The generated response content obtained in this way reflects past response content (third response content), thus improving the consistency of response content in a series of automated response processes.

[0047] [Modification 1 of the third embodiment] In the third embodiment, in step S102 (Figure 5), if it is determined that the inquiry content does not match any of the pre-configured inquiry contents, the selection unit 1243 of the dispatcher 324 selects a specific agent 125-s from agents 125-1, ..., 125-N that is suitable for the inquiry content (s∈{1, ..., N}). Here, the selection unit 1243 may select the specific agent 125-s based on the above-mentioned past answer content (third answer content representing pre-configured answer content corresponding to pre-configured inquiry content that matches a second inquiry content that is earlier than the current inquiry content). For example, the selection unit 1243 may select agent 125, which is capable of searching FAQ data corresponding to past answer content (third answer content), as the specific agent 125-s. Alternatively, for example, the selection unit 1243 may select as a specific agent 125-s an agent 125 selected using the method illustrated in the first embodiment (Examples 1 to 3 of the method for selecting a specific agent 125-s) and an agent 125 capable of searching FAQ data corresponding to past answer content (third answer content). Alternatively, for example, when the selection unit 1243 selects an agent 125 using the method illustrated in the first embodiment (Examples 1 to 3 of the method for selecting a specific agent 125-s), it may use search conditions that make it easier to select an agent 125 capable of searching FAQ data corresponding to past answer content (third answer content) (step S304'). This improves the consistency between past answer content (third answer content) and newly generated answer content using the generated AI.

[0048] [Modification 2 of the third embodiment] The dispatcher 324 of the third embodiment had a determination unit 1241, response units 3242, 3244, and a selection unit 1243. However, the determination unit 1241 of the dispatcher 324 may be replaced with the determination unit 2241 of the second embodiment. In this case, the pre-setting described in the second embodiment is performed, and step S102 is replaced with step S202 described in the second embodiment (Figure 5). Furthermore, step S304' described in Modification 1 of the third embodiment may be performed instead of step S104.

[0049] [Modification 3 of the third embodiment] In the third embodiment and its modifications 1 and 2, the past response content (third response content) referred to the response content of the FAQ data (pre-configured response content corresponding to pre-configured inquiry content that matches the past second inquiry content). However, in the third embodiment and its modifications 1 and 2, the past response content (third response content) may also be the response content based on the generated response content obtained by a specific agent 125-s for the past inquiry content (steps S105, S305).

[0050] [Hardware configuration] The functions realized by the components described herein may be implemented in a circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), CPUs (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to realize the functions described herein. A processor includes transistors and other circuits and is considered a circuitry or processing circuitry. A processor may be a programmed processor that executes a program stored in memory.

[0051] In this specification, circuitry, unit, and means are hardware programmed to perform or execute the functions described herein. Such hardware may be any hardware disclosed herein, or any hardware known to be programmed to perform or execute the functions described herein.

[0052] If the hardware is a processor that is considered to be a type of circuitry, then the circuitry, means, or unit is a combination of hardware and software used to constitute the hardware and / or processor.

[0053] For example, the terminal device 11 and the automatic response devices 12, 22, and 32 in each embodiment are devices configured by a general-purpose or dedicated computer equipped with a processor (hardware processor) such as a CPU (central processing unit) and memory such as RAM (random-access memory) and ROM (read-only memory) executing a predetermined program. That is, the terminal device 11 and the automatic response devices 12, 22, and 32 in each embodiment have, for example, processing circuits configured to implement the respective parts they each possess. This computer may have one processor and memory, or it may have multiple processors and memories. This program may be installed on the computer, or it may be pre-recorded in ROM or the like. Furthermore, some or all of the processing units may be configured using electronic circuits that realize processing functions independently, rather than electronic circuits that realize the functional configuration by loading a program, such as a CPU. Also, the electronic circuits that constitute one device may include multiple CPUs.

[0054] Figure 8 is a block diagram illustrating the hardware configuration of the terminal device 11 and the automatic response devices 12, 22, and 32 in each embodiment. As illustrated in Figure 8, the terminal device 11 and the automatic response devices 12, 22, and 32 in this example have a CPU (Central Processing Unit) 10a, an input unit 10b, an output unit 10c, a RAM (Random Access Memory) 10d, a ROM (Read Only Memory) 10e, an auxiliary storage device 10f, a communication unit 10h, and a bus 10g. The CPU 10a in this example has a control unit 10aa, an arithmetic unit 10ab, and a register 10ac, and performs various arithmetic processing according to various programs loaded into the register 10ac. The input unit 10b is an input terminal, keyboard, mouse, touch panel, etc., into which data is input. The output unit 10c is an output terminal, display, etc., into which data is output. The communication unit 10h is a LAN card, etc., controlled by the CPU 10a which has loaded a predetermined program. Furthermore, RAM 10d is an SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), etc., and has a program area 10da where a predetermined program is stored and a data area 10db where various data is stored. Furthermore, auxiliary storage device 10f is, for example, a hard disk, MO (Magneto-Optical disc), semiconductor memory, etc., and has a program area 10fa where a predetermined program is stored and a data area 10fb where various data is stored. Furthermore, bus 10g connects CPU 10a, input unit 10b, output unit 10c, RAM 10d, ROM 10e, communication unit 10h, and auxiliary storage device 10f so that information can be exchanged. CPU 10a writes the program stored in the program area 10fa of auxiliary storage device 10f to the program area 10da of RAM 10d according to the loaded OS (Operating System) program. Similarly, CPU 10a writes various data stored in the data area 10fb of auxiliary storage device 10f to the data area 10db of RAM 10d.The addresses on RAM 10d where the program and data are written are then stored in register 10ac of CPU 10a. The control unit 10aa of CPU 10a sequentially reads these addresses stored in register 10ac, reads the program and data from the area on RAM 10d indicated by the read addresses, sequentially has the arithmetic unit 10ab execute the calculations indicated by the program, and stores the calculation results in register 10ac. This configuration realizes the functional configuration of the terminal device 11 and the automatic response devices 12, 22, and 32.

[0055] The program describing this process can be recorded on a computer-readable recording medium. Examples of computer-readable recording media are non-transitory recording media. Examples of such recording media include magnetic recording devices, optical discs, magneto-optical recording media, and semiconductor memory.

[0056] Furthermore, this program may be distributed, for example, by selling, transferring, or lending portable recording media such as DVDs or CD-ROMs on which the program is recorded. Alternatively, the program may be stored in the storage device of a server computer and distributed by transferring the program from the server computer to other computers via a network.

[0057] A computer executing such a program may, for example, first store the program recorded on a portable storage medium or a program transferred from a server computer in its own storage device. Then, when processing is to be executed, the computer reads the program stored on its own storage medium and executes the processing according to the read program. Alternatively, the computer may directly read the program from the portable storage medium and execute the processing according to that program, or it may sequentially execute the processing according to the received program each time a program is transferred to it from a server computer. Furthermore, the processing may be executed by a so-called ASP (Application Service Provider) type service, where the processing function is realized only by execution instructions and result acquisition, without transferring the program from the server computer to this computer. Furthermore, the processing may be executed using a so-called SaaS (Software as a Service) type service, where a part of the server computer is made available to the user along with the program. In this form, the program includes information used for processing by an electronic computer that is equivalent to a program (data that is not a direct instruction to the computer but has the property of defining the computer's processing).

[0058] Furthermore, in this configuration, the device is configured by executing a predetermined program on a computer, but at least a part of these processes may be implemented in hardware.

[0059] [Other variations] It should be noted that the present invention is not limited to the embodiments described above. For example, in the embodiments described above, the determination units 1241 and 2241 (Figure 3) directly access the FAQ data contained in the data 1212 stored in the storage unit 120 and determine whether the inquiry content matches any of the pre-configured inquiry contents. However, in the pre-configuration, the FAQ data contained in the data 1212 may be duplicated, and the duplicated FAQ data may be stored in the storage unit 120 or other storage unit. In this case, instead of directly accessing the FAQ data contained in the data 1212, the determination units 1241 and 2241 may access the duplicated FAQ data and determine whether the inquiry content matches any of the pre-configured inquiry contents.

[0060] Furthermore, the various processes described above may not only be executed sequentially as described, but may also be executed in parallel or individually as needed, depending on the processing capacity of the device performing the processes. It goes without saying that other modifications can be made as appropriate without departing from the spirit of the present invention.

[0061] [Note] To summarize the above points, it is as follows: [Note 1] A determination unit that determines whether the inquiry content matches any of the pre-set inquiry content, A first response unit outputs a first response content representing a pre-configured response content corresponding to a pre-configured inquiry content that matches the aforementioned inquiry content, if the aforementioned inquiry content matches any of the pre-configured inquiry contents. An automated response device having a second response unit that, if the aforementioned inquiry does not match any of the pre-set inquiry contents, outputs a second response content based on the generated response content obtained by a specific agent using generation AI for the aforementioned inquiry content. [Note 2] The automatic response device as described in Appendix 1, The aforementioned specific agent belongs to a set of one or more agents, Each of the one or more agents has a function to obtain generated response content using at least the search results obtained by searching for data that can be searched based on the query content and the generating AI. The data searchable by at least one of the one or more agents includes information representing the pre-configured inquiry content and information representing the pre-configured response content, in an automated response device. [Note 3] An automatic response device as specified in Appendix 1 or 2, The determination unit determines whether the inquiry content matches any of the pre-set inquiry contents based on a determination criterion that represents the completeness of the pre-set response content, The frequency at which the indicator is determined to have a first value and the pre-set response content matches the inquiry content is higher than the frequency at which the indicator is determined to have a second value and the pre-set response content matches the inquiry content. An automatic response device in which the level of completeness represented by the first value is higher than the level of completeness represented by the second value. [Note 4] An automatic response device as described in Appendix 1 to 3, An automated response device in which the specific agent acquires the generated response content by using at least a third response content representing a pre-configured response content corresponding to a pre-configured inquiry content that matches a second inquiry content that is more recent than the aforementioned inquiry content, and the generating AI. [Note 5] An automatic response device as described in Appendix 1 to 4, An automated response device further comprising a selection unit that selects a specific agent based on a third response content representing a pre-configured response content corresponding to a pre-configured inquiry content that matches a second inquiry content that is more recent than the aforementioned inquiry content. [Note 6] An automatic response device as described in Appendix 1 to 5, An automated response device that includes additional information indicating that the first response content is a pre-set response content. [Note 7] An automatic response device as specified in Appendix 4 or 5, The above-mentioned inquiry content and the above-mentioned second inquiry content were transmitted from the same terminal device, according to the automated response system. [Note 8] An automatic response device as specified in Appendix 4 or 5, An automated response system in which the aforementioned inquiry content and the aforementioned second inquiry content are transmitted from a terminal device of a user with the same attributes. [Note 9] An automated response method performed by an automated response device, A determination step in which the determination unit determines whether the inquiry content matches any of the pre-set inquiry content, If the aforementioned inquiry content matches any of the pre-configured inquiry content, the first response unit outputs a first response content representing a pre-configured response content corresponding to the pre-configured inquiry content that matches the aforementioned inquiry content (first response step), If the aforementioned inquiry does not match any of the pre-configured inquiry contents, the second response unit outputs a second response based on the generated response content obtained by a specific agent using generation AI for the aforementioned inquiry content, in a second response step. A highly automated response method. [Note 10] A program to make a computer function as an automated response system as described in one of the appendices 1 through 9. [Explanation of symbols]

[0062] 12,22,32 Automatic response device 124,224,324 dispatchers 1241,2241 Judgment section 1242,3242 Answer part (1st answer part) 1244,3244 Answer part (2nd answer part) 1243 Selection Section

Claims

1. A determination unit that determines whether the inquiry content matches any of the pre-set inquiry content, A first response unit outputs a first response content representing a pre-configured response content corresponding to a pre-configured inquiry content that matches the aforementioned inquiry content, if the aforementioned inquiry content matches any of the pre-configured inquiry contents. An automated response device having a second response unit that, if the aforementioned inquiry does not match any of the pre-set inquiry contents, outputs a second response content based on the generated response content obtained by a specific agent using generation AI for the aforementioned inquiry content.

2. An automatic response device according to claim 1, The aforementioned specific agent belongs to a set of one or more agents, Each of the one or more agents has a function to obtain generated response content using at least the search results obtained by searching for data that it can search based on the query content and the generating AI. The data searchable by at least one of the one or more agents includes information representing the pre-configured inquiry content and information representing the pre-configured response content, in an automated response device.

3. An automatic response device according to claim 1 or 2, The determination unit determines whether the inquiry content matches any of the pre-set inquiry contents based on a determination criterion that represents the completeness of the pre-set response content, The frequency at which the indicator is determined to have a first value and the pre-set response content matches the inquiry content is higher than the frequency at which the indicator is determined to have a second value and the pre-set response content matches the inquiry content. An automatic response device in which the level of completeness represented by the first value is higher than the level of completeness represented by the second value.

4. An automatic response device according to claim 1 or 2, An automated response device in which the specific agent acquires the generated response content by using at least the generating AI and a third response content representing a pre-configured response content corresponding to a pre-configured inquiry content that matches a second inquiry content that is more recent than the aforementioned inquiry content.

5. An automatic response device according to claim 1 or 2, An automated response device further comprising a selection unit that selects a specific agent based on a third response content representing a pre-configured response content corresponding to a pre-configured inquiry content that matches a second inquiry content that is more recent than the aforementioned inquiry content.

6. A program for causing a computer to function as an automatic response device according to claim 1 or 2.

Images