Topic-generating device, topic-generating method, and topic-generating program

Abstract

To generate topics that ensure psychological safety among users belonging to a group chat. [Solution] The topic provider device is characterized by comprising: an acquisition unit that acquires logs of a group chat to which multiple users belong; an analysis unit that analyzes the conversation content of the multiple users based on the acquired logs; a generation unit that generates topics to be provided to the group using a generation model based on the psychological state of each user obtained from the results of the analysis; and an output unit that outputs the generated topics to the group chat.

Description

Technical Field

[0001] The present disclosure relates to a topic providing device, a topic providing method, and a topic providing program for promoting conversations between users.

Background Art

[0002] In recent years, in the chats used by users, there are cases where topics are provided by a generation model to facilitate communication.

[0003] In the technology where a generation model provides topics, a communication support device analyzes the conversations of users received from a terminal and identifies keywords that may form communication with another user based on the analysis results. Subsequently, the communication support device generates a dialogue for promoting communication with another user based on the identified keywords and outputs it to the user's terminal (Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] By using the prior art, it is possible to encourage users to communicate with other users. By the way, the use of group chats has been increasing year by year because the spread of remote work and the increasing opportunities for highly specialized technologies in work have led to an increase in the working styles in small groups. When using group chats for work or study, in order to improve work efficiency and group performance, the users within the group must be able to work with confidence. Therefore, it is required to provide topics that take into account the psychological safety of users using group chats.

[0006] Therefore, this disclosure proposes a topic-generating device, a topic-generating method, and a topic-generating program that can ensure psychological safety among users belonging to a group chat. [Means for solving the problem]

[0007] To solve the above problems, one form of topic provision device according to this disclosure is characterized by comprising: an acquisition unit that acquires logs of a group chat to which multiple users belong; an analysis unit that analyzes the conversation content of the multiple users based on the acquired logs; a generation unit that generates topics to be provided to the group chat using a generation model based on the psychological state of each user obtained from the results of the analysis; and an output unit that outputs the generated topics to the group chat. [Effects of the Invention]

[0008] According to one embodiment, the system generates topics that can ensure psychological safety among users belonging to a group chat. [Brief explanation of the drawing]

[0009] [Figure 1] Figure (1) shows an overview of the topic provision system according to the embodiment. [Figure 2] Figure (2) shows an overview of the topic provision system according to the embodiment. [Figure 3] Figure (1) shows an overview of the topic provision process according to the embodiment. [Figure 4] Figure (2) shows an overview of the topic provision process according to the embodiment. [Figure 5] This figure shows an example configuration of a topic-providing device according to an embodiment. [Figure 6] This figure shows an example of a conversation log information storage unit according to the embodiment. [Figure 7] Figure (1) shows an example of a member information storage unit according to the embodiment. [Figure 8]It is a diagram (2) showing an example of a member information storage unit according to an embodiment. [Figure 9] It is a diagram (1) showing an example of a team score information storage unit according to an embodiment. [Figure 10] It is a diagram (2) showing an example of a team score information storage unit according to an embodiment. [Figure 11] It is a diagram showing an example of a bot setting information storage unit according to an embodiment. [Figure 12] It is a diagram (1) showing an example of a screen in which an output unit according to an embodiment outputs a topic. [Figure 13] It is a diagram (2) showing an example of a screen in which an output unit according to an embodiment outputs a topic. [Figure 14] It is a flowchart (1) showing the procedure of a topic providing process according to an embodiment. [Figure 15] It is a flowchart (2) showing the procedure of a topic providing process according to an embodiment. [Figure 16] It is a hardware configuration diagram showing an example of a computer that realizes the functions of a topic providing device.

Mode for Carrying Out the Invention

[0010] Hereinafter, embodiments of the present disclosure will be described in detail based on the drawings. In the following embodiments, the same parts are denoted by the same reference numerals, and redundant descriptions are omitted.

[0011] (1. Embodiment) (1-1. Outline of the Topic Providing System According to the Embodiment) The topic providing process according to the present disclosure is implemented by the topic providing system 1 shown in FIG. 1. Using FIG. 1, the configuration of the topic providing system 1 according to the embodiment and the outline of the topic providing process by the topic providing system 1 will be described. FIG. 1 is a diagram (1) showing the outline of the topic providing system 1 according to the embodiment.

[0012] As shown in FIG. 1, the topic providing system 1 includes a topic providing device 100 and a user terminal 10. However, the number of the topic providing device 100 and the user terminal 10 is not limited to those shown in FIG. 1.

[0013] The topic providing device 100 is an information processing device that generates topics to be provided in group chats. For example, the topic providing device 100 is a server device, a PC (Personal Computer), a cloud server, or the like. As shown in FIG. 1, the topic providing device 100 includes a chatbot function 100a, a management function 100b, a management database 100c, and a conversation log analysis function 100d.

[0014] The chatbot function 100a is an automatic conversation program in which a computer conducts conversations on behalf of humans, and is a function in which a generation model learns the conversation logs between users by machine learning and generates topics and answers. The chatbot function 100a analyzes, for example, the conversation content in a group chat and generates topics to be provided in the group chat. Also, for example, the chatbot function 100a determines the timing for providing a topic based on the schedule information of the user stored in the management database 100c described later, and generates and provides the topic.

[0015] Management function 100b has the function of analyzing information on the status of users and groups, which will be displayed on the management dashboard described later. Management function 100b is a web application that has a survey function, a group score viewing function, and a bot management function. The survey function is a survey conducted for users using the topic provider device 100. For example, the survey function may include requests for chatbot character settings or input of user profile information. The group score viewing function is a function that allows users to view a score related to the communication status of a group based on chat conversation logs and surveys. For example, the group score viewing function allows users to view a score of the relationship between users based on conversation logs. The bot management function is a function that allows users to set the character of the chatbot function 100a to be introduced into the group chat, the pattern of topic generation, the time of posting, etc.

[0016] The management database 100c is a database that manages information used for topics generated by the chatbot function 100a. The management database 100c includes conversation logs, member information, group scores, and bot settings.

[0017] The conversation log stores, for example, the content of conversations between users using the topic provider 100. The member information stores information about users using the topic provider 100. The group score stores the communication score of the group when using the topic provider 100. The bot settings store settings such as the character of the generative model that generates the topics to be provided.

[0018] The Conversation Log Analysis Function 100d is a function that analyzes the conversation logs of group chats and analyzes the state of the group. For example, the Conversation Log Analysis Function 100d analyzes the number of participants per conversation, the frequency of conversations, the posting frequency of each user, and trending words.

[0019] User terminal 10 is a terminal device used by a user. User terminal 10 includes user terminal 10a and a management dashboard.

[0020] User terminal 10a is a terminal device owned individually by users belonging to the group chat. A user is a member of the organization or company belonging to the group chat. For example, a user can enter the group chat from their own user terminal 10a and post messages.

[0021] The management terminal 10b is a terminal device used by the administrator. The administrator is a person who manages multiple group chats. The management terminal 10b can, for example, view what kind of conversations are taking place within the group chat. In the following, when it is not necessary to distinguish between user terminal 10a and management terminal 10b, they will simply be referred to as "user terminal 10".

[0022] The user terminal 10 mentioned above is, for example, an information processing device such as a desktop PC, tablet PC, notebook PC, mobile phone, smartphone, or PDA (Personal Digital Assistant).

[0023] The topic provider 100 analyzes group chat conversations either in advance or in real time, and generates topics to provide based on the content of the analyzed conversations. The topic provider 100 then stores the generated topics in a management database and uses them when generating topics in the future. As a result, the topic provider 100 can provide topics that match the content of the conversations within the group, thereby promoting conversation among users.

[0024] Furthermore, the topic provider 100 may use the psychological safety measurement tool 100e, as shown in Figure 2, to generate topics that take into account the user's psychological safety. Figure 2 is a diagram (2) showing an overview of the topic provider system 1 according to an embodiment. As shown in Figure 2, the topic provider system 1 includes the topic provider 100, the psychological safety measurement tool 100e, and the user terminal 10.

[0025] The psychological safety measurement tool 100e is a tool for measuring the psychological safety of users belonging to a group chat. Psychological safety refers to a state in which users can confidently express their opinions and feelings to anyone in the group. The psychological safety measurement tool 100e measures users' psychological safety, for example, through a questionnaire. The topic provider device 100 may have the psychological safety measurement tool 100e installed within itself, or it may be a device that works in conjunction with a dedicated psychological safety measurement tool. In the following explanation, the psychological safety measurement tool 100e will be described assuming that it is installed externally to the topic provider device 100.

[0026] To improve teamwork among users in a group chat, the topic provider 100 measures the psychological safety of users before introducing the chatbot using the psychological safety measurement tool 100e. Subsequently, the topic provider 100 introduces the chatbot into the group chat and generates and provides topics to the group chat.

[0027] Next, the topic provider 100 analyzes the conversation logs within the group chat using the conversation log analysis function 100d to visualize the group situation. For example, the topic provider 100 analyzes the conversation logs and performs a stress analysis of the users. Then, the topic provider 100 measures the psychological safety after introducing the chatbot into the group chat using the psychological safety measurement tool 100e. This allows the topic provider 100 to measure the change in users' psychological safety before and after introducing the topic provider 100 into the group chat. Therefore, the topic provider 100 can determine whether or not the introduction of the chatbot has improved the users' psychological safety. For example, the topic provider 100 measures the change in psychological safety before and after introducing the chatbot over a period of several weeks to several months.

[0028] If the measurement results show an improvement in the user's psychological safety, the topic provider 100 determines that the topic provided to the group chat was appropriate. Conversely, if the measurement results show a decrease in the user's psychological safety, the topic provider 100 determines that the topic provided to the group chat was inappropriate or that there is a lack of understanding of the relationships between the users. In this case, the topic provider 100 changes the topic provided to the group chat, re-analyzes the relationships between the users, etc.

[0029] In the examples shown in Figures 1 and 2, the topic provider 100 generates and outputs topics to provide to the user. Conventional topic provider devices had difficulty determining when to provide the generated topic. For example, conventional topic provider devices sometimes have times when users are unable to reply due to work or school commitments. In other words, even if a conventional topic provider provides a topic to a group chat, the reply rate will decrease if the user is busy. Therefore, it is necessary to provide topics that are relevant to the current work content at the right time.

[0030] Furthermore, the use of group chats is on the rise year by year due to the increasing prevalence of remote work and the growing demand for highly specialized skills in the workplace, leading to an increase in small-group work arrangements. To improve work efficiency and group performance when using group chats, it is essential that users within the group feel safe and secure. Therefore, it is necessary to provide topics that consider the psychological safety of the users participating in the group chat.

[0031] Therefore, the topic provider 100 according to this embodiment solves the above problem by having the configuration described below. For example, the topic provider 100 acquires the work status of multiple users belonging to the same group chat, generates topics to provide to the group using a generation model based on the acquired work status, and outputs the generated topics. As a result, the topic provider 100 can provide topics that match the work status to a group chat to which multiple users belong, thereby promoting communication among users.

[0032] Furthermore, for example, the topic provider 100 can provide topics that take into account the psychological state of each user to a group chat to which multiple users belong, thereby promoting communication among users.

[0033] (1-2. Overview of Topic Provision Processing) Next, we will explain the overview of the topic provision process by the topic provision system 1 using Figures 3 and 4. Figure 3 is a diagram (1) showing an overview of the topic provision process according to the embodiment.

[0034] First, the topic provider 100 executes a schedule for providing topics to the group chat or generates a timing for posting topics (step S21). Executing a schedule means that the timing for providing topics is pre-set, and the topic provider 100 automatically provides topics to the group chat at the set time. For example, the topic provider 100 generates topics to provide to the group chat three times a day: in the morning, at noon, and in the evening. Alternatively, the topic provider 100 may change the topic provision schedule for each group.

[0035] Next, the topic provider 100, which has analyzed the timing of providing a topic, determines whether or not to respond to the user's post based on conditions such as the content of the user's post to the group chat or the content of the post entered by voice (step S22). For example, if a conversation between users is ongoing, the topic provider 100 determines not to respond to the user's post.

[0036] When responding to a post, the topic provider 100 generates a topic to offer in response to the post or conversation (step S23). When generating a topic, the topic provider 100 refers to multiple pieces of information, including each user's work status. For example, when generating a topic, the topic provider 100 refers to member profile information (step S24).

[0037] Member profile information is personal information of users belonging to a group chat. For example, member profile information includes the user's name, age, gender, contact information, and department. It also includes schedule information such as the user's working hours, meeting and business trip dates, and paid leave dates.

[0038] Furthermore, for example, the topic provider 100 refers to the conversation log when generating a topic to provide (step S25). The conversation log is a history of work-related conversations between users posted in the group chat.

[0039] Furthermore, for example, the topic provider 100 acquires external current events data when generating topics to be provided (step S26). External current events data refers to social events that occur outside the organization or company. For example, external current events data includes weather information, economic information, and international news.

[0040] Furthermore, for example, the topic provider 100 generates current emotion data (step S27). This current emotion data represents the real-time emotions of each user belonging to the group chat. For example, the topic provider 100 analyzes the conversation log to estimate the emotions of each user.

[0041] Then, the topic provider 100 outputs the topic generated based on the information from steps S24 to S27 to the group chat (step S28).

[0042] Next, we will explain the topic provision process using Figure 4. Figure 4 is a diagram (2) showing an overview of the topic provision process according to the embodiment.

[0043] The topic provider 100 executes the processes from steps S21 to S23 in Figure 3. When generating a topic to provide, the topic provider 100 refers to several additional pieces of information in addition to the information from steps S24 to S27. For example, when generating a topic to provide, the topic provider 100 refers to the group score (step S31). The group score is a score related to group communication, based on conversation logs and questionnaires.

[0044] Furthermore, for example, when generating a topic to offer, the topic provider 100 acquires a psychological safety score (step S32). This allows the topic provider 100 to generate topics tailored to each user's psychological state. The topic provider 100 then outputs the topics generated based on the information from steps S24 to S32 to the group chat (step S28).

[0045] (1-3. Configuration of the topic-providing device according to the embodiment) Next, the configuration of the topic provision device 100 that performs the topic provision processing according to the embodiment will be described. Figure 5 is a diagram showing an example of the configuration of the topic provision device 100 according to the embodiment.

[0046] As shown in Figure 5, the topic provider device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The communication unit 110 is implemented, for example, by a NIC (Network Interface Card) or a network interface controller. The communication unit 110 is connected to a network (e.g., the Internet) by wire or wireless connection and transmits and receives information with the user terminal 10, etc., via the Internet. For example, the communication unit 110 may transmit and receive information using any communication standard or technology such as Wi-Fi (registered trademark), Bluetooth (registered trademark), SIM (Subscriber Identity Module), or LPWA (Low Power Wide Area).

[0047] The memory unit 120 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or by storage devices such as hard disks and optical discs.

[0048] The memory unit 120 includes a conversation log information memory unit 121, a member information memory unit 122, a team score information memory unit 123, and a bot setting information memory unit 124. The information stored in the memory unit 120 will be explained below using Figures 6 to 11. In the examples shown in Figures 6 to 11, the information stored in the memory unit 120 may be conceptually represented as "A01," but in reality, each piece of information described later will be stored in the memory unit 120.

[0049] The conversation log information storage unit 121 stores information about conversation logs between users belonging to a group chat, which is among the information used in the topic provision processing according to the embodiment. Figure 6 is a diagram showing an example of the conversation log information storage unit 121 according to the embodiment. As shown in Figure 6, the conversation log information storage unit 121 stores the following items: "group name", "group ID", and "conversation log". "Group name" is the name assigned to each group chat. "Group ID" is an identification number assigned to each group chat. "Conversation log" is the history of conversations related to business for each group chat.

[0050] As illustrated in the example in Figure 6, the topic provider device 100 remembers that the conversation log for the group chat, whose group name is "Group A" and whose group ID is "A01", is "x01".

[0051] The member information storage unit 122 stores information about users belonging to the group chat, which is among the information used in the topic provision processing according to the embodiment. Figure 7 is a diagram (1) showing an example of the member information storage unit 122a according to the embodiment. As shown in Figure 7, the member information storage unit 122a stores the following items: "group name", "group ID", "number of members", and "group work status". "Number of members" is the number of users belonging to the group chat. "Group work status" is information about the overall work status of the group.

[0052] As illustrated in the example in Figure 7, the topic provider device 100 remembers that the group chat, whose group name is "Group A" and group ID is "A01", has "10 members", and the group's work status is "a01".

[0053] Figure 8 is a diagram (2) showing an example of the member information storage unit 122b according to the embodiment. As shown in Figure 8, the member information storage unit 122b stores the following items: "User ID", "Schedule Information", and "Individual Work Status". "User ID" is an identification number assigned to a user belonging to a specific group. "Schedule Information" is information about the user's most recent schedule. "Individual Work Status" is information about the work status of individual users belonging to the group.

[0054] As illustrated in the example in Figure 8, the topic provider 100 remembers that the user with user ID "aa01" has a schedule information of "meeting on Monday" and that the individual's work status is "aaa01".

[0055] The team score information storage unit 123 stores information about the communication status of the group chat, which is among the information used in the topic provision processing according to the embodiment. Figure 9 is Figure (1) showing an example of the team score information storage unit 123a according to the embodiment. As shown in Figure 9, the team score information storage unit 123a stores the following items: "group ID", "communication score", "stress level", "conversation frequency", and "frequently used words". "Communication score" is a score that quantifies the communication between users belonging to the group chat. For example, the communication score is evaluated on a 5-point scale. "Stress level" is a score that quantifies the stress level of the entire group chat. For example, the stress level is evaluated on a 5-point scale. "Conversation frequency" is a numerical representation of the frequency of conversations that take place in a particular group. For example, conversation frequency represents how many times a conversation takes place in a day. "Frequently used words" are words or keywords that are frequently used within a particular group.

[0056] As illustrated in the example in Figure 9, the topic provider device 100 remembers that the group with group ID "A01" has a communication score of "4", a stress level of "1.9", a conversation frequency of "5 times" per day, and a frequently occurring word "question".

[0057] Figure 10 is a diagram (2) showing an example of the team score information storage unit 123b according to the embodiment. As shown in Figure 10, the team score information storage unit 123b stores the following items: "User ID", "Psychological Safety Score", "Results of Sentiment Analysis of Conversation Content", and "Relationships between Users". The "Psychological Safety Score" is a score that quantifies the psychological safety of the users. The "Results of Sentiment Analysis of Conversation Content" is the result of analyzing the emotions of the entire group from the content of conversations between users. The "Relationships between Users" is information about the relationships between users within the group.

[0058] To illustrate with the example in Figure 10, the topic provider 100 remembers that the user with user ID "aa01" has a psychological safety score of "7.5", the sentiment analysis result of the conversation content is "calm", and the relationship between the users is "good".

[0059] The bot configuration information storage unit 124 stores the chatbot configuration information among the information used in the topic provision processing according to the embodiment. Figure 11 is a diagram showing an example of the bot configuration information storage unit 124 according to the embodiment. As shown in Figure 11, the bot configuration information storage unit 124 stores the following items: "Group ID", "Character", "Character preference", "Posting time", and "Posting frequency". "Character" stores information about the character of the chatbot to be introduced into the group. "Character" includes "Name", "Personality", and "Speech style". "Name" is the name assigned to the chatbot. "Personality" is the chatbot's personality. "Speech style" is the chatbot's speech style. "Character preference" is information on whether or not users belonging to the group have preferences regarding the chatbot's character. "Posting time" is the time when the chatbot provides topics. "Posting frequency" is the frequency with which the chatbot posts the topics it provides. For example, the posting frequency sets how many times a day a topic will be posted.

[0060] To illustrate with the example in Figure 11, the topic provider device 100 remembers that the group ID is "A01", the character's name is "D01", their personality is "E01", their speaking style is "F01", the character request is "yes", the posting time is "12:30", and the posting frequency is "3 times" per day.

[0061] Returning to Figure 5, let's continue the explanation of the control unit 130. The control unit 130 is implemented by, for example, a CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), etc., which executes a program stored inside the topic provider device 100 using RAM or the like as a working area. The control unit 130 is also a controller and is implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

[0062] As shown in Figure 5, the control unit 130 includes an acquisition unit 131, an analysis unit 132, a generation unit 133, and an output unit 134.

[0063] The acquisition unit 131 acquires logs of group chats to which multiple users belong. For example, the acquisition unit 131 acquires conversation logs between users in a group chat to which multiple users belong. Specifically, the acquisition unit 131 acquires conversation logs between users before the chatbot was introduced to the group chat. Also, for example, the acquisition unit 131 acquires conversation logs between users after the chatbot was introduced to the group chat. This allows the topic provider device 100 to acquire information to compare whether communication within the group has improved before and after the introduction of the chatbot to the group chat.

[0064] Furthermore, for example, the acquisition unit 131 acquires conversation logs before a new user joins the group chat and conversation logs after a predetermined period of time has elapsed since the new user joined the group chat. Specifically, the acquisition unit 131 acquires conversation logs before a new employee joins the group chat and conversation logs one month after the new employee joins the group chat. This allows the topic provider device 100 to compare whether or not there has been a change in communication within the group by acquiring conversation logs before and after a new user joins the group chat.

[0065] Furthermore, for example, the acquisition unit 131 acquires conversation logs before and after changing the character of the chatbot to be introduced into the group chat. This allows the topic provider device 100 to obtain information to compare whether the chatbot's character is suitable for the group.

[0066] For example, the acquisition unit 131 can acquire the results of a survey conducted among multiple users belonging to a group chat. Specifically, the acquisition unit 131 asks users to answer a survey regarding their preferences for the character of a chatbot to be introduced into the group chat, and acquires the results. Alternatively, the acquisition unit 131 can ask users to answer a survey regarding their psychological safety, and acquire the results. As a result, the topic provider device 100 can gain a detailed understanding of each user's current situation and opinions by acquiring the survey results.

[0067] The acquisition unit 131 acquires the work status of multiple users belonging to the same group chat. For example, the acquisition unit 131 acquires the work status of multiple users belonging to the same group chat from each user's user terminal 10. Work status refers to the amount of work and working hours of each user belonging to the same group chat. For example, the acquisition unit 131 acquires the work schedules of multiple users from the schedule information registered in the calendar of the user terminal 10. Also, for example, the acquisition unit 131 acquires the amount of work of the group or users from the information stored in the member information storage unit 122. Specifically, the acquisition unit 131 acquires the meeting and business trip schedules, holiday schedules, detailed working hours, etc. of each user stored in the member information storage unit 122. As a result, the topic provider device 100 can grasp the detailed schedules of users.

[0068] The analysis unit 132 analyzes the conversation content of the multiple users based on the acquired logs. For example, the analysis unit 132 analyzes the conversation content between users based on the conversation logs acquired by the acquisition unit 131. Specifically, the analysis unit 132 analyzes the conversation content before and after the chatbot is introduced into the group chat. This allows the topic provider device 100 to analyze the changes in communication within the group before and after the chatbot is introduced into the group chat. In addition, for example, the analysis unit 132 analyzes frequently occurring keywords and words from the acquired conversation logs.

[0069] The analysis unit 132 analyzes the acquired work status. For example, the analysis unit 132 analyzes the work status of each user acquired by the acquisition unit 131. Work status refers to the user's work schedule, such as daily working hours, meeting and business trip schedules, and paid leave.

[0070] The analysis unit 132 analyzes the status of multiple users' work hours based on the work status acquired by the acquisition unit 131, determining whether the users' work hours are below a threshold. For example, if a user's work hours are below a threshold, the analysis unit 132 analyzes that the user has ample time. Conversely, if a user's work hours are above a threshold, the analysis unit 132 analyzes that the user does not have ample time.

[0071] The analysis unit 132 analyzes, for example, the frequency of conversations between multiple users from the acquired conversation logs. Specifically, the analysis unit 132 analyzes from the acquired conversation logs which users have a high or low frequency of conversations with each other.

[0072] The analysis unit 132 analyzes the latest conversation content based on the acquired chat logs. For example, the analysis unit 132 acquires the content of the currently ongoing conversation and analyzes the latest conversation content. Specifically, the analysis unit 132 acquires the content of the currently ongoing conversation and analyzes when the conversation will end. This allows the topic provider device 100 to avoid interrupting communication between users by refraining from providing topics until the conversation has completely finished.

[0073] The analysis unit 132 analyzes who the user's conversation partner is. For example, the analysis unit 132 analyzes other users the user is talking to. Specifically, the analysis unit 132 analyzes from the acquired conversation logs which users the user talks to frequently and which users the user talks to infrequently. In other words, the analysis unit 132 analyzes that users who talk to each other frequently are communicating appropriately, and that users who talk to each other infrequently are not communicating properly. As a result, the topic provider device 100 can visualize users who are not communicating well with each other.

[0074] The analysis unit 132 analyzes the conversation content of multiple users and scores the relationships between those users. Specifically, the analysis unit 132 scores the relationships between users, determining that a high score indicates a good relationship and a low score indicates a weak relationship. The analysis unit 132 may also represent the scored relationships in a graph for the user to review.

[0075] The analysis unit 132 analyzes the user's psychological state from the acquired conversation log. For example, the analysis unit 132 analyzes from the conversation log whether the user's stress score is above a threshold. Specifically, if the user's stress score is above a threshold, the analysis unit 132 analyzes that the user has high stress levels. Also, if the user's stress score is below a threshold, the analysis unit 132 analyzes that the user has low stress levels.

[0076] The analysis unit 132 analyzes each user's psychological state based on the acquired psychological safety score. For example, the analysis unit 132 analyzes each user's communication score and stress level based on the psychological safety score acquired from an external psychological safety measurement tool. This allows the topic provider device 100 to appropriately analyze each user's psychological state.

[0077] The generation unit 133 generates topics to be provided to the group chat using a generation model, based on the work situation analyzed by the analysis unit 132. For example, the generation unit 133 generates topics related to work content based on the work situation analyzed by the analysis unit 132. Specifically, the generation unit 133 generates relevant topics based on the analyzed work content of each user.

[0078] The generation unit 133 generates topics to be provided to the group chat using a generation model, for example, when the users' work hours are below a threshold. For example, the generation unit 133 analyzes each user's work hours and generates topics to be provided to the group chat when the work hours are below a predetermined threshold. A state where work hours are below a threshold means that the majority of users belonging to the group chat have free time.

[0079] The generation unit 133 generates topics to be provided to the group chat using a generation model based on the psychological state of each user obtained from the analysis results. For example, the generation unit 133 generates topics to be provided to the group chat based on the measured psychological safety score of each user. A high psychological safety score indicates that each user in the group chat is in a state where they can easily participate in the conversation with peace of mind. Conversely, a low psychological safety score indicates that each user in the group chat is in a state where they cannot easily participate in the conversation with peace of mind. For example, if the measured psychological safety score is low, the generation unit 133 generates topics to improve psychological safety. Specifically, the generation unit 133 generates topics that allow new employees or users unfamiliar with their work to converse with other users with peace of mind. In this way, the topic provider device 100 can improve psychological safety within the group chat by providing topics that each user can converse with peace of mind.

[0080] The generation unit 133 changes the topics it generates based on the work status and the attributes of multiple users acquired by the acquisition unit 131. The attributes of each user include, for example, new employees and veteran employees. For example, when the majority of users belonging to the group chat have free time, the generation unit 133 generates topics tailored to new employees.

[0081] Furthermore, for example, the generation unit 133 changes the topic it generates depending on the time of day it is provided. Specifically, when providing a topic at 9:00 AM, the generation unit 133 generates the sentence, "Let's do our best today," but when providing a topic at 6:00 PM, it generates the sentence, "Thank you for your hard work today."

[0082] The generation unit 133 generates topics that match the latest conversation content analyzed by the analysis unit 132. For example, the generation unit 133 generates topics that match the content of the conversation currently in progress. Specifically, if there is a user who is not participating in the conversation currently in progress, the generation unit 133 provides topics that make it easier for that user to join the conversation. Also, for example, the generation unit 133 generates topics that match the latest conversation content based on the psychological safety score measured using a psychological safety measurement tool. Specifically, the generation unit 133 generates topics that allow a user with a low psychological safety score to participate in the conversation currently in progress.

[0083] The generation unit 133 generates topics for users whose conversation frequency is below a threshold, based on the analysis of conversations by the analysis unit 132. For example, the generation unit 133 generates topics for users whose conversation frequency is relatively low compared to other users, based on the analysis results. Specifically, since new employees often have fewer conversations with other users, the generation unit generates topics to facilitate communication between new employees and other existing employees. Also, for example, the generation unit 133 generates topics for users who only converse with specific users, based on the analysis results. Furthermore, for example, the generation unit 133 generates topics for users with low psychological safety scores to share with each other.

[0084] The generation unit 133 generates topics for conversation with users other than the conversation partner analyzed by the analysis unit 132. For example, the generation unit 133 generates topics to provide to users who have relatively few conversations with each other, based on the analysis results. Specifically, the generation unit 133 generates topics for conversation with other users for users who have only had conversations with a particular user, based on the analysis results.

[0085] The generation unit 133 generates topics to be provided to the group chat based on relationship scores. For example, the generation unit 133 modifies and generates topics to be provided to the group chat based on the relationship scores between users analyzed by the analysis unit 132. Specifically, the generation unit 133 generates topics to promote communication between users whose relationship scores are below a threshold, based on the analysis results. Also, for example, the generation unit 133 generates topics that make it easier for multiple users with low measured psychological safety scores to participate in the conversation. Specifically, the generation unit 133 generates topics to be provided to users who have low relationship scores and low measured psychological safety scores.

[0086] The output unit 134 outputs the topics generated by the generation unit 133 to the group chat. For example, the output unit 134 outputs the topics generated by the generation unit 133 to the group chat on the set days and times. Specifically, if the output unit 134 is set to provide topics at 9:00 and 12:00, it will output the topics twice, at 9:00 and 12:00. Also, for example, the output unit 134 outputs topics at times when each user has free time. This allows the topic provider device 100 to increase the response rate from users.

[0087] The output unit 134 changes the timing of outputting the generated topic. For example, the output unit 134 changes the timing so that each user outputs the topic when they have free time. Specifically, the output unit 134 outputs the generated topic to the group chat when the work time analyzed by the analysis unit 132 is below a threshold. This allows the topic provider device 100 to facilitate communication by outputting topics at a time when users are more likely to respond.

[0088] Next, we will explain examples of topics output by the chatbot using Figures 12 and 13. Figure 12 is an example of a screen showing a topic output by the output unit 134 according to the embodiment (Figure (1)). As shown in Figure 12, the output unit 134 outputs the topic generated by the generation unit 133 to the group chat.

[0089] In Figure 12, the output unit 134 displays a chat screen on the terminal screen of the user terminal 10 using the topic provision device 100 and outputs a topic related to work. Below, we will explain an example of a conversation between user a and user b. First, let's assume that the output unit 134 outputs the topic "User a, is there anything you need help with?" as the generated topic (step S301).

[0090] In response to the chatbot's topic, user a replies, "There are a few things I don't understand about the XX project." (Step S302). Next, in response to user a's post, user b replies, "I've worked on a project similar to the XX project in the past. What specifically don't you understand?" (Step S303). Next, in response to user b's post, user a replies, "The first thing is about how to write the document. How can I write it in a way that makes it easy to understand?" (Step S304). Next, in response to user a's post, user b replies, "You're asking about the document's content. I've attached some documents from a past project for reference, so let's use those documents in our meeting." (Step S305).

[0091] The output unit 134 determines when the conversation between user a and user b has ended, and if it determines that the conversation has ended, it outputs the topic "User c, do you have any advice for user a? (Step S306)" to the other user, user c.

[0092] As described above, the topic provider 100 generates and outputs topics that encourage conversation among users a, b, and c, who belong to the group chat. The topic provider 100 analyzes each user's work time and determines that users a and b have free time, then provides them with the topic in step S301. In addition, the topic provider 100 provides a topic like the one in step S306 to user c, who is not participating in the conversation between users a and b and has free time. In this way, the topic provider 100 promotes communication within the group.

[0093] Figure 13 is a diagram (2) showing an example of a screen where the output unit 134 according to the embodiment outputs a topic. In Figure 13, the output unit 134 displays a chat screen on the terminal screen of the user terminal 10 using the topic provision device 100 and outputs a topic to enhance psychological safety. Below, we will explain an example of a conversation between a new employee user d and existing employees user e and user f. First, let's assume that the output unit 134 outputs the topic "User d, have you gotten used to the new environment?" as a generated topic (step S401).

[0094] In response to the chatbot's topic, user d replies, "I'm still a little nervous." (Step S402). In response to user d's post, user e replies, "I was like that at first too." (Step S403). Then, in response to posts from users d and e, user f replies, "That's right, you just have to take your time and get used to it." (Step S404).

[0095] The output unit 134 determines when the conversation between user d, user e, and user f has ended, and when the conversation has ended, it outputs the message "User e and user f, please actively support user d! (Step S405)" to user e and user f.

[0096] The topic provider 100 generates and outputs topics that encourage conversation among users d, e, and f, who belong to the group chat. The topic provider 100 provides the topic in step S401 because user d is a new employee and their measured psychological safety score is lower than that of existing employees, users e and f. The topic provider 100 also provides topics like step S405, which encourage support for user d, because existing employees, users e and f, have high measured psychological safety scores and are accustomed to their work. In this way, the topic provider 100 can provide topics that enhance psychological safety within the group.

[0097] (1-4. Procedure for providing topics according to the embodiment) The procedure for the management process according to the above embodiment will be specifically explained using Figures 14 and 15. Figure 14 is a flowchart (1) showing the procedure for the topic provision process according to the embodiment.

[0098] As shown in Figure 14, the topic provider 100 determines whether or not it has acquired the work status of multiple users belonging to the same group chat (step S101). If it has not acquired the work status of multiple users (step S101; No), the topic provider 100 waits until it acquires the work status. On the other hand, if it has acquired the work status of multiple users (step S101; Yes), the topic provider 100 analyzes the work status of multiple users from the acquired work status (step S102).

[0099] Next, the topic provider 100, having analyzed the work situation, generates a topic to provide to the group chat based on the analyzed work situation (step S103). Then, the topic provider 100 determines whether or not it has output the generated topic to the group chat (step S104). If the generated topic has not been output to the group chat (step S104; No), the topic provider 100 waits until the group chat conversation ends. On the other hand, if the generated topic has been output to the group chat (step S104; Yes), the topic provider 100 terminates its processing.

[0100] Figure 15 is a flowchart (2) showing the procedure for topic provision processing according to the embodiment. As shown in Figure 15, the topic provision device 100 analyzes the psychological state of multiple users using a psychological safety measurement tool (step S201). Next, the topic provision device 100 determines whether or not it has obtained the log of a group chat to which multiple users belong (step S202). If the group chat log has not been obtained (step S202; No), the topic provision device 100 waits until it obtains the group chat log. On the other hand, if the group chat log has been obtained (step S202; Yes), the topic provision device 100 analyzes the conversation content of multiple users based on the obtained log (step S203).

[0101] Next, the topic provider 100 generates a topic to provide to the group based on the psychological state of each user from the analyzed results (step S204). The topic provider 100 determines whether or not the generated topic has been output to the group chat (step S205). If the topic has not been output to the group chat (step S205; No), the topic provider 100 waits until the topic is output. On the other hand, if the topic has been output to the group chat (step S205; Yes), the topic provider 100 re-analyzes the psychological state of multiple users using the psychological safety measurement tool (step S206) and then terminates the process.

[0102] (2. Other Embodiments) The processing according to the above-described embodiment may be carried out in various other forms besides those described above.

[0103] For example, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0104] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0105] Furthermore, the embodiments and modifications described above can be combined as appropriate, provided that the processing content is not inconsistent.

[0106] Furthermore, the effects described herein are merely illustrative and not limiting; other effects may also occur.

[0107] (3. Effects of the topic-generating device related to this disclosure) As described above, the topic provision device (topic provision device 100 in the embodiment) according to this disclosure comprises an acquisition unit (acquisition unit 131 in the embodiment), an analysis unit (analysis unit 132 in the embodiment), a generation unit (generation unit 133 in the embodiment), and an output unit (output unit 134 in the embodiment). The acquisition unit acquires the work status of multiple users belonging to the same group chat. The analysis unit analyzes the acquired work status. The generation unit generates topics to be provided to the group chat using a generation model based on the analyzed work status. The output unit outputs the generated topics to the group chat.

[0108] In this way, the topic provider acquires the work status of multiple users belonging to the same group chat from each user's terminal (user terminal 10 in this embodiment), analyzes the acquired work status, generates topics to provide based on the analyzed work status, and outputs the generated topics to each user's terminal. As a result, the topic provider can facilitate communication among users belonging to the group chat.

[0109] Furthermore, the topic provider analyzes the work status of multiple users based on the work status acquired by the acquisition unit, determining when their work time is below a threshold. This allows the topic provider to analyze and identify periods of free time within a user's work schedule, enabling it to provide topics at times when users are more likely to respond.

[0110] Furthermore, the topic provider changes the topics it generates based on the work status acquired by the acquisition unit and the attributes of multiple users. This allows the topic provider to generate topics tailored to each user's work status and attributes, thereby providing topics that are stress-free for each user. Additionally, the topic provider can provide topics tailored to each user by generating topics that match each user's work status and attributes.

[0111] Furthermore, the topic provider device changes the timing at which it outputs generated topics. By adjusting the timing of topic output, the topic provider device can output topics at times when users are more likely to respond. Also, by adjusting the timing of topic output, the topic provider device can prevent outputting topics while users are in the middle of a conversation, thus avoiding interrupting user conversations.

[0112] As described above, the topic provision device according to this disclosure (topic provision device 100 in the embodiment) comprises an acquisition unit (acquisition unit 131 in the embodiment), an analysis unit (analysis unit 132 in the embodiment), a generation unit (generation unit 133 in the embodiment), and an output unit (output unit 134 in the embodiment). The acquisition unit acquires logs of a group chat to which multiple users belong. The analysis unit analyzes the conversation content of the multiple users based on the acquired logs. The generation unit generates topics to be provided to the group chat using a generation model based on the psychological state of each user obtained from the analysis results. The output unit outputs the generated topics to the group chat.

[0113] In this way, the topic provider can acquire conversation logs from a group chat to which multiple users belong, analyze the acquired conversation logs, and generate and output topics to be provided to the group chat based on the analyzed psychological state of each user, thereby providing topics that are appropriate to the psychological state of the users.

[0114] Furthermore, the topic provider analyzes the latest conversation content based on the acquired chat logs and generates topics that match the analyzed latest conversation content. In this way, the topic provider can generate topics that match the currently ongoing conversation content.

[0115] Furthermore, the topic provider analyzes the frequency of conversations among multiple users and generates topics for users whose conversation frequency is below a threshold. This allows the topic provider to generate topics for conversations between users who have relatively low conversation frequencies.

[0116] Furthermore, the topic provider analyzes who the user's conversation partner is and generates topics for conversation with users other than the analyzed conversation partner. This allows the topic provider to generate topics for conversations between users who have relatively low conversation frequency, thereby promoting broader communication among users.

[0117] Furthermore, the topic provider analyzes the conversation content of multiple users, scores the relationships between those users, and generates topics to provide to the group chat based on those relationship scores. This allows the topic provider to change the topics it generates based on the relationship scores. In addition, the topic provider can generate topics to encourage conversations between users with low relationship scores.

[0118] (4. Hardware Configuration) The information devices such as the topic-providing device 100 according to the embodiment described above are realized by a computer 1000 having a configuration such as that shown in Figure 16. The topic-providing device 100 according to the embodiment will be described below as an example. Figure 16 is a hardware configuration diagram showing an example of a computer 1000 that realizes the functions of the topic-providing device 100. The computer 1000 has a CPU 1100, RAM 1200, ROM (Read Only Memory) 1300, HDD (Hard Disk Drive) 1400, a communication interface 1500, and an input / output interface 1600. The various parts of the computer 1000 are connected by a bus 1050.

[0119] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, and controls various parts. For example, the CPU 1100 loads the programs stored in the ROM 1300 or HDD 1400 into the RAM 1200 and executes processing corresponding to the various programs.

[0120] ROM1300 stores boot programs such as the BIOS (Basic Input Output System) executed by CPU1100 when computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0121] HDD1400 is a computer-readable recording medium that non-temporarily records programs executed by CPU1100 and data used by such programs. Specifically, HDD1400 is a recording medium that records a program that performs the calculation processing related to this disclosure, which is an example of program data 1450.

[0122] The communication interface 1500 is an interface for the computer 1000 to connect to an external network 1550 (e.g., the Internet). For example, the CPU 1100 can receive data from other devices or transmit data it generates to other devices via the communication interface 1500.

[0123] The input / output interface 1600 is an interface for connecting the input / output device 1650 and the computer 1000. For example, the CPU 1100 receives data from input devices such as a keyboard or mouse via the input / output interface 1600. The CPU 1100 also transmits data to output devices such as a display, speaker, or printer via the input / output interface 1600. The input / output interface 1600 may also function as a media interface for reading programs recorded on a predetermined recording medium (media). Examples of media include optical recording media such as DVDs (Digital Versatile Discs) and PDs (Phase Change Rewritable Disks), magneto-optical recording media such as MOs (Magneto-Optical Disks), tape media, magnetic recording media, or semiconductor memory.

[0124] For example, when the computer 1000 functions as a topic provider 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes functions such as the control unit 130 by executing a management program loaded on the RAM 1200. The HDD 1400 stores programs that perform the management processing according to this disclosure, as well as data in the storage unit 120. The CPU 1100 reads and executes program data 1450 from the HDD 1400, but as another example, these programs may be obtained from other devices via an external network 1550.

[0125] Although embodiments of the present application have been described in detail based on the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention. [Explanation of Symbols]

[0126] 10 User terminals 100 Topic providing device 110 Communications Department 120 Storage section 121 Conversation log information storage unit 122 Member Information Storage Unit 123 Team Score Information Storage Unit 124 Bot configuration information storage unit 130 Control Unit 131 Acquisition Department 132 Analysis Department 133 Generation part 134 Output section

Claims

1. A unit for acquiring logs of group chats to which multiple users belong, Based on the acquired logs, an analysis unit analyzes the conversation content of the multiple users, Based on the psychological state of each user obtained from the results of the above analysis, a generation unit generates topics to be provided to the group chat using a generative model, An output unit that outputs the generated topic to the group chat, A topic-generating device characterized by being equipped with the following features.

2. The aforementioned analysis unit, Based on the acquired chat logs, the latest conversation content is analyzed, The generating unit is It generates topics that match the latest analyzed conversation content. The topic-generating device according to feature 1.

3. The aforementioned analysis unit, Analyze the frequency of conversations between multiple users, The generating unit is For users whose conversation frequency is below a threshold among the analyzed conversations, a topic is generated. The topic-generating device according to feature 1.

4. The aforementioned analysis unit, Analyze who the other user is in the conversation with, The generating unit is To generate a topic for conversation with users other than the user who was the other party in the analyzed conversation, The topic-generating device according to feature 1.

5. The aforementioned analysis unit, The conversation content of multiple users is analyzed, and the relationship between those users is scored. The generating unit is Based on the aforementioned relationship score, generate topics to provide to the group chat. The topic-generating device according to feature 1.

6. Computers Obtain the logs of a group chat to which multiple users belong, Based on the acquired logs, the conversation content of the multiple users is analyzed, Based on the psychological state of each user obtained from the above analysis, a generative model is used to generate topics to be provided to the group chat. Output the generated topic to the aforementioned group chat. A method for providing topics, characterized by including the following.

7. Computers, A unit for acquiring logs of group chats to which multiple users belong, Based on the acquired logs, an analysis unit analyzes the conversation content of the multiple users, Based on the psychological state of each user obtained from the results of the above analysis, a generation unit generates topics to be provided to the group chat using a generative model, An output unit that outputs the generated topic to the group chat, A topic-generating program characterized by functioning as a topic-generating device equipped with the necessary components.

Images