Work support system, work support method, and program

Abstract

To improve the convenience for a user who uses a communication tool for supporting work.SOLUTION: A posted data acquisition unit (101) of a work support system (1) acquires posted data indicating a post that has been made on a communication tool that supports work. A mention determination unit (102) determines, based on the posted data, whether a mention has been made on an artificial intelligence (AI). A prompt data acquisition unit (103) acquires prompt data indicating a prompt provided in advance. An answer data generation unit (104) causes, when it is determined that the mention has been made, the AI to generate answer data indicating an answer to the post based on the posted data and the prompt data. An answer display unit (105) displays the answer on the communication tool based on the answer data.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present disclosure relates to a business support system, a business support method, and a program. [Background technology]

[0002] Conventionally, technologies have been considered for using AI (Artificial Intelligence) such as large-scale language models to assist users in their work. For example, Patent Literature 1 describes an interactive chatbot that searches a collection of sentences that has accumulated multiple sentences to be searched based on a character string entered by a user on an interactive screen for the user to chat with a chatbot, which is an example of AI, and obtains one or more search result sentences that match the character string and presents them to the user. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-170598 Summary of the Invention [Problem to be solved by the invention]

[0004] However, with the technology of Patent Document 1, when a user tries to use a chatbot to search for a collection of texts while communicating with other users using a communication tool that supports their work, the user needs to open a dedicated dialogue screen separate from the communication tool and chat with the chatbot. For this reason, the technology of Patent Document 1 does not sufficiently improve user convenience.

[0005] One of the purposes of the present disclosure is to improve the convenience of users in using communication tools that support business operations. [Means for solving the problem]

[0006] A business support system according to one aspect of the present disclosure includes a post data acquisition unit that acquires post data indicating a post made on a communication tool that supports business operations; a mention determination unit that determines whether a mention has been made to an AI (Artificial Intelligence) based on the post data; a prompt data acquisition unit that acquires prompt data indicating a pre-prepared prompt; a response data generation unit that, when it is determined that the mention has been made, causes the AI ​​to generate response data indicating a response to the post based on the post data and the prompt data; and a response display unit that displays the response on the communication tool based on the response data. [Effects of the Invention]

[0007] According to the present disclosure, it is possible to improve the convenience of users in using communication tools that support business. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 illustrates an example of a hardware configuration of a business support system. [Figure 2] FIG. 10 is a diagram illustrating an example of a communication tool screen. [Figure 3] FIG. 10 is a diagram showing an example of a state in which a user makes a new post. [Figure 4] FIG. 2 is a diagram illustrating an example of functions realized by the business support system. [Figure 5] FIG. 2 is a diagram illustrating an example of a user database. [Figure 6] FIG. 10 is a diagram illustrating an example of a communication tool database. [Figure 7] FIG. 10 is a diagram illustrating an example of prompt data. [Figure 8] FIG. 2 is a diagram illustrating an example of processing executed in the business support system. [Figure 9] FIG. 10 is a diagram illustrating an example of functions realized by a modified example of a business support system. DETAILED DESCRIPTION OF THE INVENTION

[0009] [1. Hardware configuration] An example of an embodiment of a business support system, a business support method, and a program according to the present disclosure will be described. Fig. 1 is a diagram showing an example of a hardware configuration of a business support system. For example, the business support system 1 includes a server 10 and a user terminal 20. Each of the server 10 and the user terminal 20 is connected to a network N such as the Internet or a LAN.

[0010] The server 10 is a server computer. For example, the server 10 includes a control unit 11, a storage unit 12, and a communication unit 13. The control unit 11 includes at least one processor. The storage unit 12 includes at least one of a volatile memory such as RAM and a non-volatile memory such as a flash memory. The communication unit 13 includes at least one of a communication interface for wired communication and a communication interface for wireless communication.

[0011] The user terminal 20 is a user's computer. For example, the user terminal 20 is a personal computer, a tablet terminal, a smartphone, or a wearable terminal. For example, the user terminal 20 includes a control unit 21, a memory unit 22, a communication unit 23, an operation unit 24, and a display unit 25. The hardware configurations of the control unit 21, the memory unit 22, and the communication unit 23 may be similar to those of the control unit 11, the memory unit 12, and the communication unit 13, respectively. The operation unit 24 is an input device such as a mouse or a touch panel. The display unit 25 is a liquid crystal or organic EL display.

[0012] The programs stored in the storage units 12, 22 may be supplied via the network N. The hardware configurations of the server 10 and the user terminal 20 are not limited to the example in FIG. 1. For example, at least one of the server 10 and the user terminal 20 may include at least one of a reading unit (e.g., a memory card slot) that reads a computer-readable information storage medium and an input / output unit (e.g., a USB terminal) for direct connection to an external device. The programs stored in the information storage medium may be supplied to at least one of the server 10 and the user terminal 20 via at least one of the reading unit and the input / output unit.

[0013] Furthermore, the business support system 1 may include at least one computer. The computers included in the business support system 1 are not limited to the example in FIG. 1. For example, the business support system 1 may include only the server 10. In this case, the user terminal 20 exists outside the business support system 1. The business support system 1 may include the server 10 and other server computers.

[0014] [2. Overview of the business support system] In this embodiment, the business support system 1 has various functions for supporting the user's business. For example, the business support system 1 may provide the user with cloud-based or on-premise groupware. The business support system 1 may also provide the user with a service that supports business, although it is not classified as groupware. The business support system 1 of this embodiment has, as one of the above functions, a communication tool that supports business. The communication tool is a tool that allows the user to communicate with other users regarding business matters.

[0015] In this embodiment, threads are described as an example of a communication tool, but other types of communication tools may also be used. The communication tool is not limited to threads. For example, the communication tool may be a bulletin board, chat, a comment function in the business support system 1, a messaging app, SNS, SMS, or other tools that are not classified as threads. In a case where a type of database called an app with a comment function is provided to a user, the comment function of the app may correspond to the communication tool.

[0016] For example, a user cooperates with other users on a team in an organization such as a company to carry out work. The user communicates with other users on the same team regarding work using a communication tool provided by the work support system 1. For example, when a user logs in to the work support system 1 and selects a communication tool, the user terminal 20 displays a communication tool screen showing the communication tool on the display unit 25. In this embodiment, a case where the communication tool screen is displayed on a browser of the user terminal 20 is taken as an example.

[0017] FIG. 2 is a diagram showing an example of a communication tool screen. As shown in FIG. 2, the user terminal 20 displays on the communication tool screen SC the content of a thread selected by a user from among multiple threads created in the business support system 1. For example, the communication tool screen SC displays posts made to the thread in chronological order. A post is the content of communication. A post can also be called a message or text. For example, a post is a character string, a file, a link, or a combination of these. A user can post a new post to a thread or post a reply to an existing post.

[0018] FIG. 3 is a diagram showing an example of how a user makes a new post. As shown in the upper part of FIG. 3, the user inputs a new post into an input form F10. In this embodiment, the user can mention another user by specifying that user. A mention is a notification to another user. For example, a user can mention another user by entering a specific symbol (e.g., @) followed by the other user's information (e.g., the other user's name). The mention is displayed on the screen of the other user. The mechanism for mentioning may be similar to a known mechanism. For example, an email indicating that the mentioned user has been mentioned may be sent to the mentioned user.

[0019] In this embodiment, a user can mention an AI (Artificial Intelligence) in the same way as mentioning other users. AI is a program with artificial intelligence that supports a user's work. There are various definitions of AI, and the AI ​​in this embodiment may be AI defined by various known definitions. The AI ​​may be AI called generative AI or conversational AI. For example, the AI ​​may be a large-scale language model, a machine learning model not classified as a large-scale language model, a program called a bot, or other programs. There are various definitions of machine learning, and the machine learning in this embodiment may be machine learning defined by various known definitions. The machine learning may be any of supervised learning, semi-supervised learning, and unsupervised learning.

[0020] In this embodiment, a case where a large-scale language model corresponds to an AI is taken as an example. For example, a user can mention an AI by entering a specific symbol (e.g., @) followed by a character string indicating the AI ​​(e.g., "bot") instead of information about another user. When an AI is mentioned, the server 10 causes the AI ​​to generate answer data indicating a response to the user's post based on post data indicating the user's post (in the example of the communication tool screen SC at the top of FIG. 3, the sentence "I will be explaining popular programming languages ​​to my manager at next week's regular meeting. Please tell me about popular programming languages.") and prompt data indicating a prompt prepared in advance on the business support system 1 side.

[0021] For example, the server 10 transmits data to the user terminal 20 for displaying the answer indicated by the answer data generated by the AI ​​on the communication tool screen SC. As shown in the lower part of Figure 3, the user terminal 20 displays the AI's answer (in the example at the bottom of Figure 3, the sentence "Trends in programming languages ​​are constantly changing, but as of 2023, several languages ​​are attracting attention. For example, the following five languages ​​are...") on the communication tool screen SC based on the data received from the server 10. The AI's answer may mention the user who posted the answer. In the example at the bottom of Figure 3, the user is mentioned by the string "@user U1" included in the AI's answer.

[0022] As described above, in the business support system 1, when a user posts a message mentioning an AI in a communication tool, the AI ​​generates response data indicating a response in response to the user's post. The AI's response is displayed in the communication tool. This eliminates the need for the user to switch back and forth between the communication tool and another tool for causing the AI ​​to generate a response, thereby enabling the business support system 1 to improve user convenience. Details of the business support system 1 will be described below.

[0023] [3. Functions realized by the business support system] FIG. 4 is a diagram showing an example of functions realized by the business support system 1. As shown in FIG.

[0024] [3-1. Functions realized by the server] For example, the server 10 includes a data storage unit 100, a post data acquisition unit 101, a mention determination unit 102, a prompt data acquisition unit 103, a response data generation unit 104, and a response display unit 105. Each of the components in the data storage unit 100 is realized by the memory unit 12. Each of the post data acquisition unit 101, the mention determination unit 102, the prompt data acquisition unit 103, the response data generation unit 104, and the response display unit 105 is realized by the control unit 11.

[0025] [Data storage section] The data storage unit 100 stores data necessary for supporting business operations. For example, the data storage unit 100 stores a user database DB1 in which various types of user data are stored, a communication tool database DB2 in which various types of data on communication tools are stored, and prompt data PD indicative of prepared prompts.

[0026] Fig. 5 is a diagram showing an example of the user database DB1. For example, the user database DB1 stores a user ID, a password, user name data indicating the user's name, and user attribute data. Any data may be stored in the user database DB1. The data stored in the user database DB1 is not limited to the example of Fig. 5. For example, the user database DB1 may store organization data indicating the organization to which the user belongs.

[0027] The user ID is an example of user identification data used by the server 10 to identify a user. When the user ID is used for logging in, the user ID is sometimes called a user account. The user identification data may be other data such as an email address or a phone number. The password is used for authentication at the time of logging in. The name indicated by the user name data is used to mention the user. Other data such as the user ID may be used to mention the user. The user attribute data is data indicating the attributes of the user. Details of the user attribute data will be described in the modified example below.

[0028] FIG. 6 is a diagram showing an example of the communication tool database DB2. For example, the communication tool database DB2 stores a thread ID, which is an ID for identifying a thread, and post data indicating posts to the thread. Any data may be stored in the communication tool database DB2. The data stored in the communication tool database DB2 is not limited to the example of FIG. 6. For example, the communication tool database DB2 may store the user ID of the user who posted, the posting date and time, the mentioned user or AI, and data on reactions to the post.

[0029] For example, when a user creates a new thread, the server 10 issues a thread ID for that thread. When a new post is made to that thread, the post data acquisition unit 101, described below, associates post data indicating the new post with the issued thread ID and stores the data in the communication tool database DB2. The post data may include not only text indicating the content of the post, but also other data such as files included in the post. In this embodiment, response data indicating a response from an AI is also stored in the communication tool database DB2 as a type of post data, but the response data may also be stored in other databases.

[0030] FIG. 7 is a diagram showing an example of prompt data PD. The prompt data PD indicates a default prompt prepared in advance. A prompt is an instruction (input) to an AI. A post indicated by posting data is also a type of prompt. In this embodiment, an example is given in which the prompt data PD is prepared by an operator managing the business support system 1. The prompt data PD may be prepared by a user or by an administrator in the organization to which the user belongs. In the example of prompt data PD in FIG. 7, the post indicated by the posting data is inserted after "User:".

[0031] The data stored in the data storage unit 100 is not limited to the above example. The data storage unit 100 can store any data. For example, the data storage unit 100 may store actual data of an AI. The actual data of an AI includes a program that indicates processing such as calculation of embedded representations, and parameters referenced by the program. The actual data of an AI may also include knowledge data that indicates knowledge accumulated in the AI. It is assumed that learning of a large-scale language model, which is an example of an AI, has been completed, but the parameters can be adjusted by re-learning such as fine tuning. The AI ​​may be stored in a computer other than the server 10. In this case, the server 10 communicates with the other computer to use the AI.

[0032] When the AI ​​is a large-scale language model, as in this embodiment, the AI ​​may be any large-scale language model. For example, the AI ​​may be a Transformer-based model such as ChatGPT or BERT, or other models that are not Transformer-based. The AI ​​may be a machine learning model that is not classified as a large-scale language model. For example, the AI ​​may be a model such as a generative adversarial network, a neural network, or a support vector machine. The AI ​​may be a program that is not classified as a machine learning model, such as a large-scale language model. For example, the AI ​​may be a rule-based AI or a program that analyzes user posts using natural language processing such as morphological analysis.

[0033] For example, the data storage unit 100 may store programs and data for each function of the business support system 1. The data storage unit 100 may store data such as communication tool programs, data referenced by the programs, and data for displaying the communication tool screen SC (e.g., HTML data). The data storage unit 100 may also store data for functions other than communication tools (e.g., a schedule management function for managing user schedules, a file management function for managing user files, or an email management function for managing shared emails).

[0034] [Posted Data Acquisition Department] The post data acquisition unit 101 acquires post data indicating posts made using a communication tool that supports business operations. For example, the post data acquisition unit 101 acquires post data from the user terminal 20. In the example of the communication tool in FIGS. 2 and 3, the post data indicates posts entered by the user in the input form F10. The post data indicates a character string entered by the user. The post data may indicate information other than a character string. For example, the post data may indicate an emoticon, file, image, video, link, or other information specified by the user.

[0035] For example, when a user makes a new post in a thread, the user terminal 20 transmits the thread ID of the thread in which the new post was made and post data indicating the new post to the server 10. The post data acquisition unit 101 acquires the thread ID and post data from the user terminal 20. The post data acquisition unit 101 associates the thread ID and post data acquired from the user terminal 20 with each other and stores them in the communication tool database DB2. The post data acquisition unit 101 can acquire the post data stored in the communication tool database DB2 at any timing.

[0036] [Mention Determination Section] The mention determination unit 102 determines whether or not a mention has been made to the AI ​​based on the posting data. In this embodiment, the mention determination unit 102 determines whether or not a mention has been made to the AI ​​based on the posting data using the same method as for mentioning users who use a communication tool. The mention method may be a known method. In this embodiment, an example is given in which a mention is made by inputting a specific symbol (for example, @) followed by a character string indicating the user's name. The mention may be made using a method that uses other symbols, or may be made using an image such as a button without inputting a particular symbol.

[0037] For example, the mention determination unit 102 determines whether a specific character string indicating a mention of an AI (e.g., bot) is input after a specific symbol in a post indicated by the post data. If the mention determination unit 102 determines that a specific character string is not input after the specific symbol, it does not determine that a mention of an AI has been made. If the mention determination unit 102 determines that a specific character string is input after the specific symbol, it determines that a mention of an AI has been made.

[0038] The mention determination unit 102 may make a determination according to the mention method defined by the communication tool. For example, if a mention of an AI is made by selecting a specific image or text displayed on the communication tool screen SC, the mention determination unit 102 may determine whether a mention of the AI ​​has been made by determining whether the specific image or text has been selected. The mention determination unit 102 may determine whether a mention of the AI ​​has been made by determining whether information in a specific format indicating a mention of the AI ​​is included in the posted data.

[0039] [Prompt data acquisition section] The prompt data acquiring unit 103 acquires the prompt data PD. In this embodiment, the prompt data PD is stored in the data storage unit 100, so the prompt data acquiring unit 103 acquires the prompt data PD from the data storage unit 100. The prompt data PD may be stored in a computer or an external information storage medium other than the server 10. In this case, the prompt data acquiring unit 103 acquires the prompt data PD from the other computer or the external information storage medium.

[0040] The prompt data PD may indicate any content other than a default prompt. For example, the prompt data PD may indicate the name of the organization to which the user belongs, the type of data such as a plug-in available to the AI, the content of the task to be performed by the AI, the format of the answer data, or a combination of these. The prompt data PD may indicate not only a character string indicating the default prompt, but also a character string indicating the insertion position of the post indicated by the post data ("User:" in the example of FIG. 7).

[0041] Note that prompt data PD may be prepared for each thread. In this case, the data storage unit 100 stores a thread ID and prompt data PD in association with each other. The prompt data acquisition unit 103 acquires prompt data PD associated with the thread ID of the thread in which the post indicated by the post data was made. Prompt data PD may be prepared for each user or organization, rather than for each thread. In this case, the data storage unit 100 stores a user ID or organization ID in association with the prompt data PD. The prompt data acquisition unit 103 acquires prompt data PD associated with the user ID or organization ID of the user who made the post indicated by the post data.

[0042] [Response data generation section] When it is determined that a mention has been made, the response data generation unit 104 causes the AI ​​to generate response data indicating a response to the post based on the post data and the prompt data PD. The response data generation unit 104 inputs each of the post data and the prompt data PD to the AI. The response data generation unit 104 may insert the post indicated by the post data into the prompt indicated by the prompt data PD and then input the prompt to the AI. The response data generation unit 104 acquires the response data output from the AI.

[0043] The processing of the AI ​​after input to the AI ​​may be publicly known processing. When the AI ​​is a large-scale language model as in this embodiment, the AI ​​divides its input into tokens based on learned parameters and calculates the embedding of each token based on the learned parameters. A token is part of the input to the AI. A token includes at least one character. The AI ​​predicts the next token as needed based on the sequence of the embedded representations of the tokens and outputs answer data indicating an answer according to the sequence of the embedded representations. This series of processing is executed by a program included in the actual data of the AI. This series of processing may be publicly known processing adopted in interactive AI, etc.

[0044] For example, if the AI ​​is a machine learning model that is not classified as a large-scale language model, the AI ​​may calculate embedded representations without particularly dividing tokens and output answer data that indicates an answer corresponding to the embedded representations. If the AI ​​is a program that is not a machine learning model, the AI ​​may determine whether or not its input matches predetermined rules without calculating embedded representations, and output answer data that indicates an answer corresponding to the result of the determination. This series of processes is also performed by a program included in the AI's actual data.

[0045] The answer data may be in any format as long as it indicates an answer from the AI. For example, the answer data may indicate a text answer from the AI. If the AI ​​generates an image as an answer, the answer data may indicate the image generated by the AI. If the AI ​​obtains a file managed by the business support system 1 as an answer, the answer data may indicate the file. The answer data may indicate a combination of two or more of text, an image, and a file.

[0046] Furthermore, the AI ​​can process any task based on user posts. In the example of Figure 3, the task corresponds to the AI ​​providing information to the user. The task may be any task that can be processed by a publicly known AI. For example, the task may be searching for information stored in a database managed by the business support system 1, searching for information managed in an internet encyclopedia, machine translation, creating a summary, generating text, generating images, proofreading posts, or other tasks. The AI ​​may determine the task that it should process without the user specifying a task.

[0047] [Answer display area] The answer display unit 105 displays the answer on the communication tool based on the answer data. Displaying the answer on the communication tool by the answer display unit 105 means that the answer display unit 105 executes processing for displaying the answer on the communication tool screen SC. In this embodiment, the answer display unit 105 is realized by the server 10, and therefore the answer display unit 105 displays the AI's answer on the communication tool by transmitting display data for displaying the answer on the communication tool to the user terminal 20. The display data may be the same as the answer data, or may be separate data generated from the answer data.

[0048] The display data may be in any format. It may be data necessary for displaying some information on the communication tool screen SC. For example, when the communication tool screen SC is displayed in a browser, the display data may be data in a markup language such as HTML. When the communication tool screen SC is displayed by an application dedicated to the business support system 1, the display data may be data (e.g., image data or text data) necessary for displaying some information in the application. In the example of FIG. 3, the answer display unit 105 displays an answer from the AI ​​as a reply to the post made to the user. The user's post is also displayed on the communication tool screen SC by processing of the server 10.

[0049] [3-2. Functions implemented on user devices] For example, the user terminal 20 includes a data storage unit 200, a display control unit 201, and an operation reception unit 202. The data storage unit 200 is realized by the storage unit 22. The display control unit 201 and the operation reception unit 202 are each realized by the control unit 21.

[0050] [Data storage section] The data storage unit 200 stores data for business support. For example, the data storage unit 200 stores a browser for displaying various screens of the business support system 1. For example, the data storage unit 200 stores a program dedicated to the business support system 1.

[0051] [Display control section] The display control unit 201 causes the display unit 25 to display various screens in the business support system 1. For example, the display control unit 201 causes the display unit 25 to display various screens such as a communication tool screen SC based on data received from the server 10.

[0052] [Operation reception section] The operation reception unit 202 receives various operations in the business support system 1. For example, it receives operations on various screens such as the communication tool screen SC. Data indicating the operation content received by the operation reception unit 202 is transmitted to the server 10 as appropriate.

[0053] [4. Processing performed by the business support system] Fig. 8 is a diagram showing an example of processing executed in the business support system 1. The processing in Fig. 8 is executed by the control units 11 and 21 executing programs stored in the storage units 12 and 22, respectively. The processing in Fig. 8 is executed when a user logs in to the business support system 1 and then selects an arbitrary thread in a communication tool.

[0054] Based on the communication tool database DB2, the server 10 executes processing between the server 10 and the user terminal 20 to display a communication tool screen SC indicating a thread selected by the user (S1). In S1, the server 10 generates display data (e.g., HTML data) for the communication tool screen SC based on the posting data of existing posts in the thread and transmits it to the user terminal 20. Based on the display data, the user terminal 20 displays the communication tool screen SC as shown in FIG. 2 on the display unit 25. The display data also indicates the thread ID of the thread to be displayed.

[0055] The user terminal 20 accepts input of a post to the input form F10 (S2). In S2, the user can make a post that includes the character string "@bot" to mention the AI, or can make a post that does not include the character string. When the user performs an operation to send a post, the user terminal 20 sends the thread ID of the currently displayed thread and post data indicating the post entered by the user to the server 10 (S3). The server 10 acquires the thread ID and post data from the user terminal 20 (S4).

[0056] The server 10 determines whether or not a mention has been made to the AI ​​based on the posted data (S5). In S5, the server 10 determines whether or not the character string "@bot" for mentioning the AI ​​is included in the posted data. If it is determined in S5 that a mention has not been made to the AI ​​(S5:N), the server 10 associates the thread ID and the posted data and stores them in the communication tool database DB2 (S6), and the process ends. In this case, no response data is generated by the AI. The process in S6 is the same as the process performed when posting in a known thread. If a mention has been made to another user, the server 10 executes a process for mentioning the other user.

[0057] If it is determined in S5 that the AI ​​has been mentioned (S5: Y), the server 10 acquires pre-stored prompt data PD (S7). The server 10 causes the AI ​​to generate answer data based on the post data and the prompt data PD (S8). In S8, the server 10 inserts the character string of the post indicated by the post data into a predetermined position indicated by the prompt data PD, and then provides input to the AI. The AI ​​performs processing such as token division and embedded expression calculation, and outputs answer data. The server 10 acquires the answer data output from the AI.

[0058] The server 10 associates the thread ID and the posting data and stores them in the communication tool database DB2 (S9). In S9, the server 10 stores the posting data of the post entered by the user and the response data generated by the AI ​​in the communication tool database DB2. The server 10 executes processing with the user terminal 20 to display the response on the communication tool based on the response data (S10), and this processing ends. In S10, the server 10 generates display data for a communication tool screen SC showing the post entered by the user and the response from the AI, and transmits it to the user terminal 20. The user terminal 20 displays the post entered by the user and the response from the AI ​​on the communication tool screen SC based on the display data.

[0059] [5. Summary of embodiments] The business support system 1 of this embodiment determines whether or not a mention has been made to the AI ​​based on the posted data. When it is determined that a mention has been made to the AI, the business support system 1 causes the AI ​​to generate answer data based on the posted data and the prompt data PD. The business support system 1 causes the communication tool to output an answer based on the answer data. This eliminates the need for the user to switch between the communication tool and another tool for causing the AI ​​to generate an answer, so the business support system 1 can improve the convenience of users who use the communication tool. For example, a user can ask a question to the AI ​​on the communication tool while communicating with other users on the communication tool, so the business support system 1 can improve the user's work efficiency.

[0060] Furthermore, the business support system 1 determines whether or not a mention has been made to the AI ​​in the same way as a mention to a user who uses a communication tool, based on the posting data. This allows the user to mention the AI ​​in the same way as when mentioning other users, so the business support system 1 can effectively improve user convenience.

[0061] [6. Modifications] The present disclosure is not limited to the above-described embodiments, and may be modified as appropriate without departing from the spirit of the present disclosure.

[0062] 9 is a diagram showing an example of functions realized in a modified business support system 1. As shown in FIG. 9, the modified example described below realizes a location data acquisition unit 106, a business data acquisition unit 107, a other posted data acquisition unit 108, a user attribute data acquisition unit 109, a task identification unit 110, and an AI identification unit 111. Each of the location data acquisition unit 106, the business data acquisition unit 107, the other posted data acquisition unit 108, the user attribute data acquisition unit 109, the task identification unit 110, and the AI ​​identification unit 111 is realized by the control unit 11.

[0063] [6-1. Variation 1] For example, a user can post to any location in a communication tool. A location in a communication tool can also be referred to as a group that collectively manages multiple posts. If the communication tool is a thread, the location corresponds to the thread in which the post was made among multiple threads created in the communication tool. If the communication tool is a chat, the location corresponds to the chat room in which the post was made among chat rooms created in the communication tool. Similarly, for other communication tools, a group that collectively manages multiple posts may correspond to a location. Since the content of the response that the AI ​​should make may change depending on the location where the post was made, the AI ​​may also take into account the location where the post was made when generating response data.

[0064] The business support system 1 of the first modified example includes a location data acquisition unit 106. The location data acquisition unit 106 acquires location data indicating a location where a post was made using a communication tool. For example, the location data acquisition unit 106 acquires location data from a user terminal 20. When a user inputs a post at a certain location, the user terminal 20 transmits the location data indicating the location along with the post data. The location data may be included as part of the post data. If the communication tool is a thread, the location data is a thread ID. If the server 10 can identify a thread by the name of the thread, the location data may indicate the name of the thread. The location data may indicate an individual tree within the thread rather than a thread. In this case, the tree corresponds to the location.

[0065] The location data may be data indicating a location according to the communication tool. For example, if the communication tool is chat, the location data may be an ID for identifying the chat room. Similarly, for other communication tools, the location data may be data that allows the server 10 to identify the location where the post was made. The location data acquisition unit 106 may acquire this location data from the user terminal 20. If location data indicating the location where the user is currently browsing is recorded in the data storage unit 100, the location data acquisition unit 106 may acquire the location data from the data storage unit 100.

[0066] The answer data generation unit 104 of variant example 1 causes the AI ​​to generate answer data further based on the location data. For example, the answer data generation unit 104 inputs the location data as a prompt to the AI. The AI ​​calculates the embedded expression based not only on the posting data and prompt data PD but also on the location data. The AI ​​outputs answer data indicating an answer corresponding to the embedded expression. The answer data generation unit 104 acquires the answer data output from the AI. The process by which the answer display unit 105 outputs an answer based on the answer data may be the same as in the embodiment.

[0067] The method by which the response data generation unit 104 acquires response data based on location data is not limited to the above example. The response data generation unit 104 may use the location data in some way to cause the AI ​​to generate response data. For example, if prompt data PD corresponding to the location data is prepared, the response data generation unit 104 may acquire prompt data PD associated with the location data and cause the AI ​​to generate response data based on the posted data and the prompt data PD. As another example, if there are multiple AIs and AIs corresponding to the location data are prepared, the response data generation unit 104 may select an AI associated with the location data and cause the AI ​​to generate response data based on the posted data and the prompt data PD.

[0068] The business support system 1 of the first modification causes the AI ​​to generate response data further based on the location data. This enables the business support system 1 to output a response according to the location where the post was made. The business support system 1 makes it easier for the user to output the response desired. For example, even within the same organization, the appropriate content of a response to a post in a thread from the development department may differ from that of a response to a post in a thread from the sales department. Even if the posts are similar, it may be better for the response from the development department to include content related to development. It may be better for the response from the sales department to include content related to sales. The AI ​​of the first modification can generate response data according to these departments, thereby improving the accuracy of the responses from the AI.

[0069] [6-2. Variation 2] For example, business data related to the user's business is registered in the business support system 1. The business data is data for business support. For example, the business data may be a file uploaded by the user, a database registered in the business support system 1, a field value in the database, or other data. This business data may be taken into account by AI. The data storage unit 100 of variant example 2 stores business data. For example, the data storage unit 100 stores, for each organization to which a user belongs, business data related to the business in that organization.

[0070] The business support system 1 of the second modification includes a business data acquisition unit 107. The business data acquisition unit 107 acquires business data related to the business of the user who posted. For example, when a user posts to a thread, the business data acquisition unit 107 acquires a file registered in the thread as business data. When a user posts using a comment function of an application, the business data acquisition unit 107 acquires data of a field of the application as business data. When a user posts to a specific record of an application, the business data acquisition unit 107 acquires data of the record as business data. Since these business data are pre-stored in the data storage unit 100, the business data acquisition unit 107 may acquire the business data by referring to the data storage unit 100.

[0071] The answer data generation unit 104 of variant example 2 causes the AI ​​to generate answer data further based on business data. For example, the answer data generation unit 104 inputs business data as a prompt to the AI. The AI ​​calculates embedded expressions based not only on the posting data and prompt data PD but also on the business data. The AI ​​outputs answer data indicating an answer corresponding to the embedded expression. The answer data generation unit 104 acquires the answer data output from the AI. The process by which the answer display unit 105 outputs an answer based on the answer data may be the same as in the embodiment.

[0072] The method by which the response data generation unit 104 acquires response data based on business data is not limited to the above example. The response data generation unit 104 may use the business data in some way to cause the AI ​​to generate response data. For example, if prompt data PD corresponding to the business data is prepared, the response data generation unit 104 may acquire prompt data PD associated with the business data and cause the AI ​​to generate response data based on the posted data and the prompt data PD. As another example, if there are multiple AIs and AIs corresponding to the business data are prepared, the response data generation unit 104 may select an AI associated with the business data and cause the AI ​​to generate response data based on the posted data and the prompt data PD.

[0073] The business support system 1 of the second modification causes the AI ​​to generate answer data further based on the business data. This enables the business support system 1 to output answers that correspond to the user's business. The business support system 1 makes it easier for the user to output answers that they desire. For example, if the AI ​​generates answer data taking into account the contents of the files registered in the thread, the AI ​​can generate answer data that corresponds to the content related to the business up to the user's post. If the AI ​​generates answer data taking into account the data registered in the app's record, the AI ​​can generate answer data that corresponds to the content of the record posted by the user using the app's comment function. As a result, the business support system 1 can display answers that are closely related to the business related to the user's post on the communication tool screen SC.

[0074] [6-3. Variation 3] For example, in a communication tool, not only posts made by a user but also other existing posts may exist. Taking a thread as an example as in the embodiment, a thread in which a user makes a post may have other existing posts. Similarly, in other communication tools such as chat, at the time a user makes a post, other existing posts may already exist. Since other posts may contain content related to the post made by the user (for example, a business topic related to the post made by the user), the AI ​​may generate answer data taking other posts into consideration.

[0075] The business support system 1 of the third modified example includes an other posted data acquisition unit 108. The other posted data acquisition unit 108 acquires other posted data indicating other posts different from the post. The other posts are existing posts. For example, when a user replies to a post, the post from which the reply originates corresponds to the other post. When a user posts to a thread, the existing posts posted to that thread correspond to the other posts. The other posted data is stored in the communication tool database DB2.

[0076] For example, the other posted data acquisition unit 108 refers to the communication tool database DB2 to acquire other posted data. Similarly, when another communication tool such as chat is adopted, the other posted data acquisition unit 108 acquires other posted data indicating other existing posts. The other posted data acquisition unit 108 may acquire other posted data for the entire past period, or may acquire other posted data for a recent predetermined period (e.g., one week). The other posted data acquisition unit 108 may acquire other posted data for the same location as the location where the user made the post.

[0077] The answer data generation unit 104 of variant example 3 causes the AI ​​to generate answer data further based on other posted data. For example, the answer data generation unit 104 inputs other posted data as a prompt to the AI. The AI ​​calculates embedded expressions based not only on the posted data and prompt data PD but also on the other posted data. The AI ​​outputs answer data indicating an answer corresponding to the embedded expression. The answer data generation unit 104 acquires the answer data output from the AI. The process by which the answer display unit 105 outputs an answer based on the answer data may be the same as in the embodiment.

[0078] The method by which the response data generation unit 104 acquires response data based on other posted data is not limited to the above example. The response data generation unit 104 may use the other posted data in some way to cause the AI ​​to generate response data. For example, if prompt data PD corresponding to the other posted data is prepared, the response data generation unit 104 may acquire prompt data PD associated with the other posted data and cause the AI ​​to generate response data based on the posted data and the prompt data PD. As another example, if there are multiple AIs and an AI corresponding to the other posted data is prepared, the response data generation unit 104 may select an AI associated with the other posted data and cause the AI ​​to generate response data based on the posted data and the prompt data PD.

[0079] The business support system 1 of the third modification causes the AI ​​to generate answer data further based on other posted data. This enables the business support system 1 to output answers that correspond not only to the post made by the user but also to other posts in the communication tool. The business support system 1 makes it easier for the user to output the answer desired. For example, other posts may indicate the process leading up to the user's post. By generating answer data based on other posted data, the AI ​​can generate answer data that corresponds to the process leading up to the user's post, thereby increasing the accuracy of answers from the AI. For example, even if the post made by the user does not indicate the process leading up to the post, the AI ​​can generate answer data that corresponds to the process up to that point.

[0080] [6-4. Variation 4] For example, the other post data acquisition unit 108 of the third modified example may acquire other post data of other posts in the same tree as the post indicated by the post data acquired by the post data acquisition unit 101, among multiple other posts made on a communication tool. A tree is a collection of posts in which replies are made one after another starting from a certain post. A tree is a collection of the starting post and reply posts made based on that post. A tree can also be said to be a collection of posts that show the same topic.

[0081] It is assumed that a tree ID capable of identifying a tree is stored in the communication tool database DB2. For example, when a user makes a new post that does not quote another post, the server 10 generates a new tree ID and stores it in the communication tool database DB2. When a reply is made to the post, the server 10 associates the tree ID with the posting data of the reply post. As a result, the original post and the reply post belong to the same tree. Other communication tools, such as chat, may also support a similar tree structure.

[0082] For example, when a user posts to a certain tree, the other posted data acquisition unit 108 acquires the tree ID of the tree to which the user posted. The other posted data acquisition unit 108 acquires other posted data associated with the tree ID from the other posted data stored in the communication tool database DB2. When a user posts to a certain tree in a certain thread, the other posted data acquisition unit 108 acquires other posted data of other posts belonging to that tree. The other posted data acquisition unit 108 does not acquire other posted data of other posts belonging to other trees in the thread. The answer data generation unit 104 uses to acquire answer data is other posted data of other posts belonging to the same tree, which is different from the third modification, but the method of generating answer data based on other posted data may be the same as the third modification.

[0083] The business support system 1 of the fourth modification acquires other post data of other posts in the same tree as the post from among multiple other posts made on a communication tool. This enables the business support system 1 to acquire answer data after taking into consideration the topics that led to the post. The business support system 1 can more easily output the answer desired by the user. For example, in a thread with many posts, there may be other posts that are completely unrelated to the post made by the user. On the other hand, other posts in the same tree as the post made by the user are likely to be related to the post made by the user, so the AI ​​can generate an answer that is closely related to the post made by the user.

[0084] [6-5. Variation 5] For example, the answer desired by a user may differ depending on the user's attributes. A user's attributes are a user classification. A user's attributes may be the type of work the user is responsible for. For example, a user's attributes may be the industry of the organization to which the user belongs, the user's department within the organization, the user's job title, demographic information such as the user's gender or age, or other information. For example, even within the same organization, a user who belongs to the development department and a user who belongs to the sales department may desire different answers. Another example is that a user with a managerial position may desire different answers from a user without a managerial position. For this reason, the AI ​​may generate answer data that takes user attributes into consideration.

[0085] The business support system 1 of the fifth modification includes a user attribute data acquisition unit 109. The user attribute data acquisition unit 109 acquires user attribute data indicating the attributes of the user who posted. In the fifth modification, since the user attribute data is stored in the user database DB1, the user attribute data acquisition unit 109 acquires the user attribute data from the user database DB1. The user attribute data may indicate any attribute, including the examples described above. The user may specify the attribute indicated by their own user attribute data. An administrator of the organization to which the user belongs may specify the user's attribute. The user attribute data may be stored in a computer other than the server 10 or an external information storage medium. The user attribute data acquisition unit 109 may acquire the user attribute data from another computer or an external information storage medium.

[0086] The answer data generation unit 104 of variant example 5 generates answer data further based on user attribute data. For example, the answer data generation unit 104 inputs user attribute data as a prompt to the AI. The AI ​​calculates embedded expressions based not only on the posting data and prompt data PD but also on the user attribute data. The AI ​​outputs answer data indicating an answer corresponding to the embedded expression. The answer data generation unit 104 acquires the answer data output from the AI. The process by which the answer display unit 105 outputs an answer based on the answer data may be the same as in the embodiment.

[0087] The method by which the response data generation unit 104 acquires response data based on user attribute data is not limited to the above example. The response data generation unit 104 may use the user attribute data in some way to cause the AI ​​to generate response data. For example, if prompt data PD corresponding to the user attribute data is prepared, the response data generation unit 104 may acquire prompt data PD associated with the user attribute data and cause the AI ​​to generate response data based on the posted data and the prompt data PD. As another example, if there are multiple AIs and the AIs are prepared according to the user attribute data, the response data generation unit 104 may select an AI associated with the user attribute data and cause the AI ​​to generate response data based on the posted data and the prompt data PD.

[0088] The business support system 1 of the fifth modification generates answer data further based on user attribute data. This allows the business support system 1 to output answers that are in accordance with not only the post made by the user but also the attributes of the user who made the post. The business support system 1 can more easily output answers that the user desires. For example, if the user attribute data indicates the user's department, the AI ​​can generate answer data that is in accordance with the content of the work that the department of the user who made the post is responsible for. If the user attribute data indicates the user's job title, the AI ​​can generate answer data that is in accordance with the job title of the user who made the post.

[0089] [6-6. Variation 6] For example, the AI ​​may generate response data by referring not only to its own accumulated knowledge but also to a database managed by the business support system 1. The data storage unit 100 of the sixth modification stores various databases in which the business data described in the third modification is stored. For example, the databases store business data indicating company regulations, business data indicating an internal dictionary, business data indicating schedules, business data indicating emails, business data indicating various information related to customers, or other business data. These databases are referenced and updated by functions of the business support system 1 other than the communication tool.

[0090] The answer data generation unit 104 of variant example 6 generates answer data by having the AI ​​refer to at least one of multiple databases managed by the business support system 1. For example, the answer data generation unit 104 inputs post data and prompt data PD to the AI. The AI ​​calculates embedded expressions based on the input and selects at least one of the multiple databases based on the embedded expressions. The AI ​​outputs answer data including business data stored in the selected database. The answer data indicates all or part of the content indicated by the business data. The AI ​​may modify the content of the business data and generate answer data indicating the modified content. The process by which the answer display unit 105 outputs an answer based on the answer data may be the same as in the embodiment.

[0091] In the business support system 1 of the sixth modification, the AI ​​generates answer data by referencing at least one of a plurality of databases managed by the business support system 1. This enables the business support system 1 to output answer data related to the user's business. For example, when a user posts a question related to company rules, the AI ​​generates answer data by referencing the company rules database, thereby enabling the AI ​​to generate answer data according to the user's post. Which database the AI ​​should refer to for each type of post is specified by an embedded expression calculated by the AI. The AI ​​can generate appropriate answer data based on the database according to the content of the post.

[0092] [6-7. Variation 7] For example, in addition to mentioning the AI, the user may also specify a task that the AI ​​should perform. In Variation 7, mentioning the AI ​​is performed by entering "@bot" as in the embodiment. The task that the AI ​​should perform is specified by the user entering a character string indicating the task after a specific symbol (e.g., #) that is different from the mention of the AI.

[0093] The business support system 1 of the seventh modification includes a task identification unit 110. The task identification unit 110 identifies a task specified by the user who posted. For example, the task identification unit 110 acquires task data indicating the task specified by the user. The task data indicates a character string of the task specified by the user. In the seventh modification, a character string indicating the content of the task is input after the specific symbol "#", so the task data indicates the specific symbol "#" and the character string. For example, if the task is for AI to machine translate a post into English, the task data indicates "#Translate to English". If the task is for AI to create a summary of the posts in the thread, the task data indicates "#Create Summary".

[0094] For example, a user may input a string such as "#translate into English" to specify a task of machine-translating the user's post into English. When the user specifies the task and posts, the user terminal 20 transmits the post data to the server 10. The post data acquisition unit 101 of the server 10 acquires the post data in the same manner as in the embodiment. The data indicating the task may be separate data from the post data.

[0095] For example, the task identification unit 110 determines whether the post indicated by the posting data contains a specific symbol "#" for specifying a task. If the task identification unit 110 determines that the post does not contain the specific symbol "#", it does not identify a task. In this case, answer data is generated by AI in the same manner as in the embodiment. If the task identification unit 110 determines that the post contains the specific symbol "#", it identifies the character string "Translate into English" after the specific symbol "#" as a task.

[0096] Note that tasks that can be specified by a user may be predetermined. In this case, the task identification unit 110 may determine whether a specific symbol "#" and a character string defined as a task that can be specified by a user are included in a post, and may acquire task data based on the result of the determination. A symbol other than "#" may be used to specify a task. The specific symbol for specifying a task may be any symbol different from the symbol for mentioning. The specific symbol for specifying a task may be the same symbol as the symbol for mentioning (for example, @). In this case, if the character string following the symbol is not the user's name, the task identification unit 110 may acquire task data by regarding the character string as indicating a task.

[0097] Furthermore, a specific symbol may not be used to specify a task. In this case, when a post containing "@bot," which indicates a mention of an AI, is made, the task identification unit 110 may perform linguistic analysis of the text of the post to identify the task and acquire task data indicating the task. For example, when tasks that can be specified by a user are predetermined, the task identification unit 110 may determine whether the text of the post contains a character string indicating the predetermined task and acquire task data based on the result of the determination. The task identification unit 110 may acquire task data based on an AI for determining the task, which is prepared separately from the AI ​​that generates the answer data. Alternatively, for example, information for specifying a task (e.g., an image or text) may be displayed on the communication tool screen SC. The user may specify a task by selecting at least one of the pieces of information.

[0098] The answer data generation unit 104 of variant example 7 causes the AI ​​to generate answer data further based on the task data. For example, the answer data generation unit 104 inputs the task data as a prompt to the AI. The AI ​​calculates the embedded expression based not only on the posting data and prompt data PD but also on the task data. The AI ​​outputs answer data indicating an answer corresponding to the embedded expression. The answer data generation unit 104 acquires the answer data output from the AI. The process by which the answer display unit 105 outputs an answer based on the answer data may be the same as in the embodiment.

[0099] The method by which the response data generation unit 104 acquires response data based on task data is not limited to the above example. The response data generation unit 104 may use the task data in some way to cause an AI to generate response data. For example, if prompt data PD corresponding to the task data is prepared, the response data generation unit 104 may acquire prompt data PD associated with the task data and cause the AI ​​to generate response data based on the posted data and the prompt data PD. As another example, if there are multiple AIs and an AI corresponding to the task data is prepared, the response data generation unit 104 may select an AI associated with the task data and cause the AI ​​to generate response data based on the posted data and the prompt data PD.

[0100] The business support system 1 of the seventh modification example causes the AI ​​to generate answer data based on a task specified by the user. This allows the business support system 1 to output answers that correspond not only to the post made by the user but also to the task specified by the user who made the post. The business support system 1 can more easily output answers that the user desires. For example, by the user specifying the task that the user wants the AI ​​to perform, the business support system 1 can cause the AI ​​to perform the task that the user desires and generate answer data, thereby improving the accuracy of answers from the AI.

[0101] [6-8. Variation 8] For example, the business support system 1 may be provided with multiple AIs. Each AI is trained with training data specialized for a specific task. For example, each AI may be re-trained, including fine tuning, based on training data specialized for a specific task. In variant example 8, three AIs are provided: AI1 specialized for machine translation, AI2 specialized for summarizing, and AI3 specialized for research. The number of AIs may be any number. For example, there may be two AIs, or there may be four or more AIs.

[0102] For example, AI1 has learned training data including training sentences written in a specific language (e.g., Japanese) and the sentences translated into another language different from the specific language (e.g., English). AI2 has learned training data including training sentences and summaries of the training sentences. AI3 has learned training data such as data from an Internet encyclopedia or data from an encyclopedia created by an organization to which the user belongs. AI1 to AI3 in a trained state are stored in data storage unit 100.

[0103] For example, the user may not only mention the AI, but also specify the AI ​​that will generate answer data. In Variation 8, as in the embodiment, the user mentions the AI ​​by inputting a specific symbol "@" followed by a character string indicating the mention of the AI. However, the character string indicates one of multiple AIs. When the user specifies AI1, the character string becomes "bot1". When the user specifies AI2, the character string becomes "bot2". When the user specifies AI3, the character string becomes "bot3". Correspondence data indicating which character string corresponds to which AI is stored in advance in the data storage unit 100.

[0104] The business support system 1 of variant example 8 includes an AI identification unit 111. The AI ​​identification unit 111 identifies the AI ​​specified by the user who posted from among multiple AIs. The AI ​​identification unit 111 analyzes the posted data based on the correspondence data to identify the AI ​​specified by the user. For example, when it is determined that a mention has been made to an AI, the AI ​​identification unit 111 identifies the AI ​​specified by the user by referring to the last number in the string of the mention. If the string of the mention is "@bot1", the AI ​​identification unit 111 identifies AI1. If the string of the mention is "@bot2", the AI ​​identification unit 111 identifies AI2. If the string of the mention is "@bot3", the AI ​​identification unit 111 identifies AI3.

[0105] Note that a specific symbol other than a mention may be used to specify an AI. In this case, the user may specify an AI by entering information indicating the AI ​​to be made to process a task after another symbol (e.g., %), in addition to "@bot" indicating a mention of the AI. For example, information for specifying an AI (e.g., an image or text) may be displayed on the communication tool screen SC. The user may specify an AI by selecting at least one of the multiple pieces of information. The user may specify multiple AIs.

[0106] The response data generation unit 104 of Modification 8 causes an AI designated by a user from among a plurality of AIs to generate response data. For example, when AI1 is identified by the AI ​​identification unit 111, the response data generation unit 104 causes AI1 from among AI1 to AI3 to generate response data. When AI2 is identified by the AI ​​identification unit 111, the response data generation unit 104 causes AI2 from among AI1 to AI3 to generate response data. When AI3 is identified by the AI ​​identification unit 111, the response data generation unit 104 causes AI3 from among AI1 to AI3 to generate response data. Although this modification differs from the embodiment in that one of a plurality of AIs generates response data, the process by which the AI ​​generates response data may be the same as in the embodiment. The process by which the answer display unit 105 outputs an answer based on the response data may also be the same as in the embodiment.

[0107] The business support system 1 of the eighth modification example causes an AI designated by a user from among a plurality of AIs to generate answer data. This allows the business support system 1 to generate answer data from the AI ​​designated by the user, not just from posts made by the user. The business support system 1 makes it easier for the user to output answers desired by the user. For example, by allowing a user to designate an AI from among a plurality of AIs that they would like to process a task, the business support system 1 can cause the AI ​​desired by the user to process the task and generate answer data, thereby improving the accuracy of answers from the AI.

[0108] [6-9. Other variations] For example, two or more of Modifications 1 to 8 may be combined. The answer data generation unit 104 may cause the AI ​​to generate answer data based only on the posted data, without based on the prompt data PD. In this case, the business assistance system 1 does not include the prompt data acquisition unit 103. A business assistance system 1 that does not include the prompt data acquisition unit 103 can also solve the problem of improving user convenience, since the user can cause the AI ​​to generate answer data from a communication tool such as a thread. Therefore, a business assistance system 1 that does not include the prompt data acquisition unit 103 is also included in the scope of the present disclosure.

[0109] For example, the functions described as being realized by the server 10 may be realized by the user terminal 20. In this case, the functions may be realized by a browser script or an application installed on the user terminal 20. For example, each function may be shared among multiple computers or may be realized by a single computer. [Explanation of symbols]

[0110] 1 Business support system, 10 Server, 11, 21 Control unit, 12, 22 Memory unit, 13, 23 Communication unit, 20 User terminal, 24 Operation unit, 25 Display unit, N Network, PD Prompt data, SC Communication tool screen, DB1 User database, DB2 Communication tool database, F10 Input form, 100 Data storage unit, 101 Post data acquisition unit, 102 Mention determination unit, 103 Prompt data acquisition unit, 104 Answer data generation unit, 105 Answer display unit, 106 Location data acquisition unit, 107 Business data acquisition unit, 108 Other posted data acquisition unit, 109 User attribute data acquisition unit, 110 Task identification unit, 111 Identification unit, 200 Data storage unit, 201 Display control unit, 202 Operation reception unit.

Claims

[Claim 1] a posting data acquisition unit that acquires posting data indicating posts made using a communication tool that supports business; a mention determination unit that determines whether or not a mention has been made to AI (Artificial Intelligence) based on the post data; a prompt data acquisition unit that acquires prompt data indicating a prepared prompt; a response data generation unit that, when it is determined that the mention has been made, causes the AI ​​to generate response data indicating a response to the post based on the post data and the prompt data; an answer display unit that displays the answer on the communication tool based on the answer data; Business support system including.

Images