Information processing apparatus, information processing method, and program

Abstract

To improve the answering accuracy of an interactive system compared with a conventional RAG system.SOLUTION: An AI system 3 executes a predetermined task for a prompt by using context information and outputs a result of the execution as generated information. A context information extraction unit 53 accesses a vector DB 71 storing a plurality of pieces of existing information with at least one of a plurality of context indices attached thereto, and extracts the context information from the vector DB 71 on the basis of at least a designated index. A UI functionalization unit 54 acquires, from a user terminal 2, a context index designated by a user together with the prompt as the designated index.SELECTED DRAWING: Figure 4

Description

[Technical Field]

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

[0002] In recent years, dialogue systems (see, for example, Patent Document 1) in which AI receives prompts and responds to them have been utilized in business activities of companies. Among dialogue systems, RAG systems have attracted attention. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7081099 Summary of the Invention [Problem to be solved by the invention]

[0004] However, conventional RAG systems do not provide high response accuracy, and there is a demand for improving the response accuracy of interactive systems. However, conventional technologies, including Patent Document 1, are unable to adequately meet this demand.

[0005] The present invention has been made in view of the above circumstances, and aims to improve the response accuracy of a dialogue system compared to conventional RAG systems. [Means for solving the problem]

[0006] In order to achieve the above object, an information processing device according to one aspect of the present invention comprises: an interactive system communication control means for controlling communication with an interactive system that executes a predetermined task in response to a prompt using context information and outputs the execution result as generated information, the control means transmitting the prompt and the context information to the interactive system and receiving the generated information output from the interactive system in response to the transmission; a context information extraction means for accessing a predetermined database in which each of a plurality of pieces of information, each of which has at least one of a plurality of indexes attached thereto, is stored, and extracting the context information from the predetermined database based on the designated index; a UI functioning means for acquiring, from the user's terminal, an index designated by the user together with the prompt as the designated index, and for causing a predetermined UI (User Interface) to function in the user's terminal for providing, to the user's terminal, the generated information received as a response from the interactive system to the prompt and the context information extracted from the predetermined database based on the designated index; Equipped with.

[0007] An information processing method and a program according to one aspect of the present invention are a method and a program corresponding to an information processing device according to one aspect of the present invention. [Effects of the Invention]

[0008] According to the present invention, it is possible to improve the accuracy of answers given by a dialogue system compared to conventional RAG systems. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram showing an overview of a service to which an embodiment of an information processing system of the present invention is applied. [Figure 2] 2 is a diagram showing an example of the configuration of an information processing system that realizes the provision of the present service in FIG. 1, that is, an embodiment of the information processing system of the present invention. FIG. [Figure 3]3 is a block diagram showing an example of a hardware configuration of a server in the information processing system of FIG. 2. FIG. [Figure 4] 4 is a functional block diagram showing an example of a functional configuration of the server of FIG. 3 that constitutes the information processing system of FIG. 2. [Figure 5] 5 is a diagram showing a specific example of a screen of a user terminal based on a predetermined UI operated by a server having the functional configuration of FIG. 4. [Figure 6] 6 is a diagram showing a specific example of a screen of a user terminal based on a predetermined UI operated by a server having the functional configuration of FIG. 4, which is different from FIG. 5. [Figure 7] 5 and 6. FIG. 7 is a diagram showing a specific example of a screen of a user terminal based on a predetermined UI operated by a server having the functional configuration of FIG. [Figure 8] 8 is a diagram showing a specific example of a screen of a user terminal based on a predetermined UI operated by a server having the functional configuration of FIG. 4, which is different from the screens of FIGS. 5 to 7. FIG. [Figure 9] 5 is a diagram showing a specific example of a management screen of a server having the functional configuration of FIG. 4. FIG. [Figure 10] 10 is a diagram showing a specific example of a management screen of a server having the functional configuration of FIG. 4, which is a screen example different from that of FIG. 9. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0011] First, with reference to FIG. 1, an overview of a service (hereinafter referred to as "this service") to which an embodiment of an information processing system of the present invention (see FIG. 3 described later) is applied will be described. 1 and 2 are diagrams showing an outline of the present service to which an embodiment of the information processing system of the present invention is applied.

[0012] This service is a service that further improves the accuracy of answers (generated information, described below) from an AI-based interactive system 3 (hereinafter referred to as "AI system 3"). In this embodiment, RAG (Retrieval Augmented Generation) is adopted as the AI ​​system 3. RAG refers to a technology that employs large-scale language models (LLMs (Large Language Models)) as AI and improves response accuracy by combining external information retrieval when generating text (generated information, described later) using the large-scale language model. RAG is also called "retrieval augmented generation" or "retrieval augmented generation." The "external information" in this "combining the search for external information when generating text (generated information, described later) using the large-scale language model (LLM)" is hereinafter referred to as "context information."

[0013] Here, we will provide an overview of the conventional RAG that is the premise of this service. First, existing information (text information) that can be context information is vectorized. Then, a set of the existing information and the vectorized existing information (hereinafter simply referred to as a “vector”) is stored in the vector DB 71. In the example of Figure 1, not only the existing information and vector pairs but also the context index (idx) are stored in the vector DB 71. This is one of the features of this service and does not exist in "traditional RAG."

[0014] A question (text information) entered into a user terminal 2 (hereinafter referred to as "user terminal 2") is output as a prompt. The prompt is vectorized to become a prompt vector. The prompt vector is provided to the AI ​​system 3. In addition, a search is performed using the prompt vector in the vector DB 71. The vector (existing information) that matches the prompt vector is provided to the AI ​​system 3 as context information.

[0015] The AI ​​system 3 generates a response sentence (text) to the prompt using the context information, and transmits the response sentence to the user terminal 2 as generated information.

[0016] In this type of "conventional RAG," the search accuracy in the vector DB71 was poor, and the context information provided to AI System 3 was not necessarily appropriate. As a result, the accuracy of the answers given by AI System 3 was also not high. Therefore, in this service, as shown in Figure 1, not only pairs of existing information and vectors but also context indexes (idx) associated with the pairs are stored in the vector DB 71. This makes it possible to perform filtering using the context index when searching the vector DB 71, and ensures that the context information provided to the AI ​​system 3 is appropriate. As a result, the accuracy of the answers given by the AI ​​system 3 is further improved.

[0017] To enable filtering using such a context index, a process of registering the context index in the vector DB 71 is required. Therefore, the process of registering a context index in the vector DB 71 will be described below.

[0018] First, when new existing information is vectorized and registered in the vector DB 71, it is confirmed whether the context index to be assigned to the existing information corresponds to one already in use in the vector DB 71, and if there is a corresponding context index in use, it is registered as metadata in the vector DB 71 together with the existing information and the vector. As shown in the example vector DB 71 of FIG. 1, the number of context indexes associated with the existing information and vectors is not limited to one, and may be multiple.

[0019] Here, if there is no context index currently in use in the vector DB 71 that corresponds to the new existing information to be registered, in step S1, the user can operate the user terminal 2 to generate a new context index that corresponds to the new existing information. Although not shown, the context index can be registered not only based on the URL of the screen, but also based on input information (screen state) such as a form. For example, if the input information for a form includes "contract status," a context index called "existing customer inquiry" can be assigned. The existing information to which this context index is assigned indicates that address change procedure documents are referenced. On the other hand, if the input information for a form does not include "contract status," a context index of "new customer inquiry" can be assigned. It can be seen that the existing information to which this context index is assigned does not reference documents for address change procedures.

[0020] In this way, not only when a user operates the vector DB71 but also when a user screen is added to the user terminal 2 or when new existing information is added to the vector DB71, a new context index is generated as appropriate, and a new context index is assigned to the relevant person among the existing information stored in the vector DB71, thereby updating the vector DB71. For example, in step S1, screens that can refer to existing information are identified from the existing system used in the user terminal 2, a new context index is generated from the screen, and the business content and identification information of the screen are combined into a context description, and the newly generated context index and context description are registered in the context DB 72. In step S2, when new existing information is vectorized, it is checked whether it corresponds to one of the context indexes registered in the context DB 72 (whether it is information that the context description can refer to), and the corresponding context index (which is information that the context description can refer to) is associated with the pair of the new existing information and the vector and stored in the vector DB 71. If no corresponding item exists, step S1 may be repeated to newly register the corresponding context index in the context DB 72, and then the newly registered context index may be associated with a new set of existing information and a vector and stored in the vector DB 71.

[0021] By using the vector DB 71 that has been updated by steps S1 and S2, it becomes possible to realize the following response process using filtering when searching the vector DB 71. As a result, the AI ​​system 3 can use appropriate context information, and can generate more accurate generated information and provide it to the user terminal 3.

[0022] That is, in the answering process, in step S3, when the screen for which the question is to be generated is the user screen (before the question is generated), the user terminal 2 specifies a context index that can filter existing information that can be referenced from the screen information, and outputs the specified context index (hereinafter referred to as the "specified index"). In step S4, the prompt output from the user terminal 2 is vectorized to become a prompt vector. In step S5, filtering is performed as a search of the vector DB 71 based on the specified index and prompt vector, and appropriate existing information and vector pairs are extracted as context information and provided to the AI ​​system 4. That is, when the user screen for which a question is to be generated is identified on the user terminal 2, a specified index is specified from the user screen (step S3), and filtering is performed as a search of the vector DB 71 based on the specified index and the prompt vector (question) (step S4), and a pair of appropriate existing information and vector is extracted as context information and provided to the AI ​​system 4 (step S5). Then, in step S5, the AI ​​system 4 uses the provided context information to generate a response sentence (text information) to the prompt as generation information. In step S6, the AI ​​system 3 transmits the generated information to the user terminal 2.

[0023] Although not shown, in filtering for searching the vector DB 71 in the answer process, the prompt vector is not essential, and only the designated index may be used.

[0024] Next, with reference to FIG. 2, a description will be given of the configuration of an information processing system that realizes the provision of the present service shown in FIG. 1, that is, an embodiment of the information processing system of the present invention. FIG. 2 is a diagram showing an example of the configuration of an information processing system that realizes the provision of the present service shown in FIG. 1, that is, an embodiment of the information processing system of the present invention.

[0025] The information processing system shown in FIG. 2 is configured to include a server 1, a user terminal 2, and an AI system 3. The server 1, the user terminal 2, and the AI ​​system 3 are connected to each other via a network such as the Internet.

[0026] The server 1 is an information processing device managed by the service provider of this service (Fig. 1). The server 1 executes various processes to realize this service while communicating with the user terminal 2 and the AI ​​system 3 as appropriate.

[0027] The user terminal 2 is an information processing device managed by a user, and is configured as a smartphone, a tablet, a personal computer, or the like.

[0028] The AI ​​system 3 is a server or the like that operates AI that is utilized in measures implemented by user companies. In this example, as described above, it is assumed that LLM is adopted as the AI ​​model.

[0029] FIG. 3 is a block diagram showing an example of a hardware configuration of a server in the information processing system shown in FIG.

[0030] The server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an input unit 16, an output unit 17, a memory unit 18, a communication unit 19, and a drive 20.

[0031] The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 18 into the RAM 13 . The RAM 13 also stores data and the like necessary for the CPU 11 to execute various processes.

[0032] The CPU 11, ROM 12, and RAM 13 are connected to one another via a bus 14. An input / output interface 15 is also connected to this bus 14. An input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20 are connected to the input / output interface 15.

[0033] The input unit 16 is configured with, for example, a keyboard and is used to input various information. The output unit 17 is configured with a display such as a liquid crystal display, a speaker, etc., and outputs various information as images and sounds. The storage unit 18 is configured with a DRAM (Dynamic Random Access Memory) or the like, and stores various data. The communication unit 19 communicates with other devices (for example, the user terminal 2 and the AI ​​system 3 in FIG. 2) via a network including the Internet.

[0034] Removable media 30, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately attached to the drive 20. A program read from the removable media 30 by the drive 20 is installed in the storage unit 18 as needed. Furthermore, the removable medium 30 can also store various data stored in the storage unit 18 in the same manner as the storage unit 18.

[0035] Although not shown, the user terminal 2 also has the hardware configuration shown in FIG.

[0036] By cooperation of various hardware (FIG. 3) and various software that constitute the information processing system of FIG. 2, including the server 1 and the user terminal 2, various processes (e.g., each process of steps S1 to S6) for providing the present service of FIG. 1 can be executed.

[0037] FIG. 4 is a functional block diagram showing an example of the functional configuration of the server of FIG. 3 in the information processing system of FIG.

[0038] As shown in FIG. 4, in the CPU 11 of the server 1, an AI system communication control unit 51, a database management unit 52, a context information extraction unit 53, and a UI function implementation unit 54 function. The storage unit 18 of the server 1 is provided with the vector DB 71 and the context DB 72 described with reference to FIG.

[0039] The AI ​​system 3 is an interactive system that executes a predetermined task in response to a prompt using context information and outputs the execution result as generated information.

[0040] The AI ​​system communication control unit 51 of the server 1 controls communication of the AI ​​system 3 by transmitting prompt and context information to the AI ​​system 4 and receiving generated information output from the AI ​​system 3 in response to that transmission.

[0041] The context information extraction unit 53 accesses the vector DB 71, which stores each of a plurality of existing information pieces to which at least one of a plurality of context indexes is assigned, and extracts context information from the vector DB 71 based on at least the specified index. The extracted context information is provided to the AI ​​system 3 under the control of the AI ​​system communication control unit 51.

[0042] In this way, the context information extraction unit 53 executes the filtering provided by this service as a search of the vector DB 71 . Here, the information used in filtering is "at least the designated index" because other information may also be used. For example, in the example of FIG. 1, the context information extraction unit 53 extracts context information from the vector DB 71 based on the prompt (more precisely, the prompt vector) together with the designated index.

[0043] The UI function unit 54 causes a predetermined UI (User Interface) to function in the user terminal 2. This predetermined UI has a function of acquiring, as a specified index, a context index specified by a user together with a prompt from the user terminal 2. Here, the "context index specified by a user" is a broad concept that includes not only a context index directly input by a user (a context index directly specified by a user), but also a context index automatically identified from a user screen displayed on the user terminal 2 at the user's specification (a context index indirectly specified by a user). In addition, this predetermined UI has a function of providing the user terminal 3 with the prompt and generated information received as a response from the AU system 4 to the context information extracted from the vector DB 71 based on the specified index.

[0044] The database management unit 52 manages the storage and retention of the vector DB 71 and the context DB 72 . For example, the database management unit 52 assigns a predetermined context index to information that is stored or to be stored in the vector DB 71 (a set of existing information and a vector in the example of FIG. 1) and manages the information. Furthermore, for example, the database management unit 52 generates a new index based on information obtained from the user terminal 2 and registers it in the context DB 72. The database management unit 52 assigns a new index to a target to be assigned a new index among information stored or to be stored in the vector DB 71 (a set of existing information and vectors in the example of FIG. 1), and manages the new index.

[0045] Next, a specific example of a screen of the user terminal 2 based on a predetermined UI functioned by the UI function unit 54 of the server 1 will be described.

[0046] FIG. 5 shows a specific example of a screen of a user terminal based on a predetermined UI operated by a server having the functional configuration of FIG. The example screen in FIG. 5 is a screen for registering a context index appropriate for the user screen of the user terminal 2 (the background screen in FIG. 5).

[0047] FIG. 6 shows a specific example of a screen of a user terminal based on a predetermined UI operated by a server having the functional configuration of FIG. 4, which is different from the screen shown in FIG. The example screen in Figure 6 is an example screen that includes a "Context Settings" area where a text explanation of the context is entered when generating a new appropriate context index for the user screen of the user terminal 2 (the background screen in Figure 5), and a "Screen Status" area where screen elements such as forms are registered. The user can enter a description of the context in text or register a screen element such as a form, and a context index (and context information) is generated according to the content and registered in the context DB 72.

[0048] FIG. 7 shows a specific example of a screen of a user terminal based on a predetermined UI operated by a server having the functional configuration of FIG. 4, which is different from the screens shown in FIGS. The example screen in Figure 7 shows how, when the user screen is a screen related to the expense settlement of a dinner party and the expense settlement of travel expenses, and the prompt is to launch a chat appropriate to the user screen, a specified index that matches the state of the user screen in Figure 7 is specified, and the optimal chat is launched based on the generated information (answer) from AI system 2 using the specified index.

[0049] FIG. 8 shows a specific example of a screen of a user terminal based on a predetermined UI operated by a server having the functional configuration of FIG. 4, which is different from the screens shown in FIGS. The example screen in Figure 8 shows how, when the user screen is a screen related to the expense settlement of business dinners and travel expenses, as in the example in Figure 7, in the case of a prompt that requests not just a response (text information) but also materials to be referenced (for example, a prompt such as "please send me materials that can confirm the maximum hotel cost"), a specified index that matches the state of the user screen in Figure 8 is specified, and the most appropriate materials can be referenced based on the information (answer) generated by AI system 2 using the specified index.

[0050] Specific examples of screens on the user terminal 2 based on a predetermined UI functioned by the UI function unit 54 of the server 1 have been described above with reference to FIGS. Next, a specific example of the management screen of the server 1 will be described with reference to FIGS.

[0051] FIG. 9 shows a specific example of a management screen of a server having the functional configuration of FIG. The example screen in Fig. 9 is an example screen displaying a list of existing information (and vectors) stored in the vector DB 71. In the example screen in Fig. 9, a given line shows information on the given existing information (and vector), and on the left side, a context index (for example, the top line shows "expense settlement work" in the clinic) attached to the given existing information (and vector) is displayed.

[0052] FIG. 10 shows a specific example of a management screen of a server having the functional configuration of FIG. 4, and is different from the screen example of FIG. The example screen of FIG. 10 is an example screen displaying a list of learning data used when performing learning to generate or update an appropriate context index.

[0053] Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and modifications, improvements, etc. within the scope of achieving the object of the present invention are considered to be included in the present invention.

[0054] For example, in the above-mentioned service, the model of AI system 2 was LLM, but it is not limited to this, and it is sufficient to use context information when generating an answer to a prompt (generated information), and various AI models are also possible, and ultimately any interactive system, not just AI, is sufficient.

[0055] Furthermore, the system configuration shown in FIG. 2 and the hardware configurations of the server 1 and the user terminal 2 shown in FIG. 3 are merely examples for achieving the object of the present invention, and are not particularly limited.

[0056] Furthermore, the functional block diagram shown in Fig. 4 is merely an example and is not particularly limited. That is, it is sufficient if the information processing system in Fig. 2 is provided with a function that can execute the various processes described above as a whole, and the functional blocks and databases used to realize this function are not particularly limited to the example in Fig. 5.

[0057] Furthermore, the locations of the functional blocks and databases are not limited to those shown in FIG. 4 and may be arbitrary. For example, at least some of the functional blocks and databases (for example, the vector DB 7 and the context DB 72) arranged on the server 1 side may be provided in another information processing device.

[0058] The above-described series of processes can be executed by hardware or software. Furthermore, one functional block may be configured as a single piece of hardware, a single piece of software, or a combination thereof.

[0059] When a series of processes is executed by software, the programs that make up the software are installed into a computer or the like from a network or a recording medium. The computer may be a computer built on dedicated hardware. The computer may also be a computer that can execute various functions by installing various programs, such as a service provider server, a general-purpose smartphone, or a personal computer.

[0060] The recording medium containing such a program may be composed of not only a removable medium (not shown) that is distributed separately from the device main body in order to provide the program to the user, but also a recording medium that is provided to the user in a state that it is pre-installed in the device main body.

[0061] In this specification, the steps describing the program to be recorded on the recording medium include not only processes that are performed chronologically in accordance with the order, but also processes that are not necessarily performed chronologically but are performed in parallel or individually.

[0062] To sum up, the information processing device to which the present invention is applied is sufficient as long as it has the following configuration, and can take on a variety of different embodiments. That is, an information processing device to which the present invention is applied (for example, the server 1 in FIGS. 1 to 4) an interactive system communication control means (e.g., AI system communication control unit 51 in FIG. 3) that controls communication with an interactive system (e.g., AI system 3 in FIGS. 1, 2, and 4) that executes a predetermined task in response to a prompt using context information and outputs the execution result as generated information, by transmitting the prompt and the context information to the interactive system and receiving the generated information output from the interactive system in response to the transmission; a context information extraction unit (e.g., the context information extraction unit 53 in FIG. 4) that accesses a predetermined database (e.g., the vector DB 71 in FIG. 1 or FIG. 4) in which a plurality of pieces of information are stored, each of which is assigned at least one of a plurality of indexes (e.g., a "context index (denoted as idx in FIG. 1 etc.)" in the embodiment), and extracts the context information from the predetermined database based on the specified index; a UI functioning means (e.g., UI functioning unit 54 in FIG. 4) that acquires, from a user terminal (e.g., user terminal 2 in FIGS. 1, 2, and 4), an index designated by the user together with the prompt as the designated index, and causes a predetermined UI (User Interface) to function in the user terminal (e.g., see FIGS. 5 to 8) for providing the generated information received from the interactive system as a response to the prompt and the context information extracted from the predetermined database based on the designated index to the user terminal; Equipped with.

[0063] the context information extraction means extracts the context information from the predetermined database based on the specified index and the prompt; It is possible.

[0064] A database management means (for example, the database management unit 52 in FIG. 4) that assigns a predetermined index to information stored or to be stored in the predetermined database (for example, the set of existing information and vectors in FIG. 1) and manages the information. The sensor may further include:

[0065] the database management means generates a new index based on the information obtained from the terminal of the user, and assigns the new index to the information to be assigned the new index among the information stored or to be stored in the predetermined database; It is possible. [Explanation of symbols]

[0066] 1 Server, 2 User terminal, 3 AI system, 11 CPU, 12 ROM, 13 RAM, 14 Bus, 15 Input / output interface, 16 Input section, 17 Output section, 18 Storage section, 19 Communication section, 20 Drive, 30 Removable media, 51 AI system communication control section, 52 Database management section, 53 Context information extraction section, 54 UI functionality section, 71 Vector DB, 72 Context DB

Claims

1. an interactive system communication control means for controlling communication with an interactive system that executes a predetermined task in response to a prompt using context information and outputs the execution result as generated information, the control means transmitting the prompt and the context information to the interactive system and receiving the generated information output from the interactive system in response to the transmission; a context information extraction means for accessing a predetermined database in which each of a plurality of pieces of information, each of which has at least one of a plurality of indexes attached thereto, is stored, and extracting the context information from the predetermined database based on the designated index; a UI functioning means for acquiring, from the user's terminal, an index designated by the user together with the prompt as the designated index, and for causing a predetermined UI (User Interface) to function in the user's terminal for providing the generated information received from the interactive system as a response to the prompt and the context information extracted from the predetermined database based on the designated index to the user's terminal; An information processing device comprising:

2. the context information extraction means extracts the context information from the predetermined database based on the prompt together with the designated index; The information processing device according to claim 1 .

3. A database management means for managing information stored or to be stored in the predetermined database by assigning a predetermined index to the information. The information processing device according to claim 1 , further comprising:

4. the database management means generates a new index based on the information obtained from the terminal of the user, and assigns the new index to the information to be assigned the new index among the information stored or to be stored in the predetermined database; The information processing device according to claim 3 .

5. An information processing method executed by an information processing device, an interactive system communication control step of controlling communication with an interactive system that executes a predetermined task in response to a prompt using context information and outputs the execution result as generated information, the interactive system transmitting the prompt and the context information to the interactive system and receiving the generated information output from the interactive system in response to the transmission; a context information extraction step of accessing a predetermined database in which each of a plurality of pieces of information, each of which has at least one of a plurality of indexes attached thereto, is stored, and extracting the context information from the predetermined database based on the designated index; a UI functioning step of acquiring, from the user's terminal, an index designated by the user together with the prompt as the designated index, and causing a predetermined UI (User Interface) to function in the user's terminal for providing the generated information received from the interactive system as a response to the prompt and the context information extracted from the predetermined database based on the designated index to the user's terminal; An information processing method including:

6. On the computer, an interactive system communication control step of controlling communication with an interactive system that executes a predetermined task in response to a prompt using context information and outputs the execution result as generated information, the interactive system transmitting the prompt and the context information to the interactive system and receiving the generated information output from the interactive system in response to the transmission; a context information extraction step of accessing a predetermined database in which each of a plurality of pieces of information, each of which has at least one of a plurality of indexes attached thereto, is stored, and extracting the context information from the predetermined database based on the designated index; a UI functioning step of acquiring, from the user's terminal, an index designated by the user together with the prompt as the designated index, and causing a predetermined UI (User Interface) to function in the user's terminal for providing the generated information received from the interactive system as a response to the prompt and the context information extracted from the predetermined database based on the designated index to the user's terminal; A program that executes control processing including:

Images