Apparatus and method for suggesting action

Abstract

To appropriately suggest actions to participants of an interaction group.SOLUTION: An apparatus extracts feature quantities of a first participant from interaction data including text messages posted by multiple participants of a first interaction group. The apparatus calculates a predetermined score of the first participant based on at least a part of the feature quantities of the first participant. The apparatus estimates a variation value of the predetermined score due to each improvement action candidate for improving the predetermined score based on the predetermined score. The improvement action candidate includes participation in a different interaction group from the interaction group. The apparatus determines a suggested improvement action for the first participant based on the variation value of the predetermined score.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to a technique for suggesting actions to participants in a social group. [Background technology]

[0002] From the perspective of ESG (Environment, Social and Governance) investment, companies that are investing in human capital and working on health management are attracting attention. The issue is how to reduce mental absence and presenteeism loss costs, which directly translate into corporate costs. On the other hand, in the case of working from home, it is becoming more difficult to grasp the stress and productivity of each employee.

[0003] A related technique of the present disclosure is JP 2020-57067 A (Patent Document 1). Patent Document 1 discloses a method for analyzing user communication without using a wearable device. Chats are analyzed, and scores of group activity, engagement, and condition are calculated. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-57067 A Summary of the Invention [Problem to be solved by the invention]

[0005] Patent Document 1 evaluates group activity, engagement, and condition from chat data. However, these evaluations alone cannot directly improve the stress and productivity of the subjects. Therefore, a technology that can suggest appropriate actions to group participants is desired. [Means for solving the problem]

[0006] In order to solve the above problem, one embodiment of the present invention provides a method for suggesting actions to participants in a first interaction group, comprising: a device extracting features of a first participant from interaction data including text messages posted by multiple participants in a first interaction group; the device calculating a predetermined score for the first participant based on at least a portion of the features of the first participant; the device estimating a change in the predetermined score due to each of candidate improvement actions that improve the predetermined score based on the predetermined score; the candidate improvement actions include participation in an interaction group different from the interaction group; and the device determining a proposed improvement action for the first participant based on the change in the predetermined score.

[0007] Further features related to the present invention will become apparent from the description of the present specification and the accompanying drawings. Also, aspects of the present invention may be realized and realized by the elements and combinations of various elements and aspects of the following detailed description and the appended claims.

[0008] It should be understood that the descriptions in this specification are exemplary and illustrative only and are not intended to limit the scope or application of the present invention in any manner. Effect of the Invention

[0009] According to one aspect of the present invention, actions can be appropriately suggested to participants in an interaction group. [Brief description of the drawings]

[0010] [Figure 1] 1 is a block diagram showing an example of a functional configuration of a system for supporting stress reduction and productivity improvement through chat analysis. [Diagram 2] FIG. 2 illustrates an example of a hardware configuration of a chat analysis device. [Diagram 3] 13 shows an example of the configuration of training data for an employee classification estimation model and a score estimation model. [Figure 4] 13 shows an example of the configuration of training data for a score improvement behavior estimation model. [Diagram 5]An example of the structure of an exchange group DB is shown below. [Figure 6] An example of the structure of the exchange data DB is shown below. [Figure 7] 13 shows a flowchart of an example of processing performed by a chat analysis device. [Figure 8] 13 shows an example of input data to the score improvement behavior estimation model. [Figure 9] 13 shows an example of the configuration of an improvement score table. [Figure 10A] An example of a screen displayed on an employee terminal is shown below. [Figure 10B] An example of a screen displayed on an employee terminal is shown below. [Figure 10C] An example of a screen displayed on an employee terminal is shown below. [Figure 10D] An example of a screen displayed on an employee terminal is shown below. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, the embodiments of the present specification will be described with reference to the accompanying drawings. In the accompanying drawings, functionally identical elements may be indicated by the same numbers. Note that the accompanying drawings show specific embodiments and implementation examples according to the principles of the present invention, but these are for understanding the present invention and are by no means used to interpret the present invention in a limiting manner.

[0012] In the present embodiment, the present invention has been described in sufficient detail for those skilled in the art to practice the present invention, but it should be understood that other implementations and forms are possible, and that changes in configuration and structure and substitutions of various elements are possible without departing from the scope and spirit of the technical ideas of the present invention. Therefore, the following description should not be interpreted as being limited thereto.

[0013] Additionally, the embodiments herein may be implemented in software running on a general purpose computer, in specialized hardware, or a combination of software and hardware, as described below.

[0014] When describing each process in the embodiments of this specification with "each processing unit as a program" as the subject (acting entity), the program is executed by a processor (CPU, etc.) to perform the defined process using memory and a communication port (communication control device), so the description may be given with the processor as the subject. <System configuration>

[0015] FIG. 1 is a block diagram showing an example of the functional configuration of a system for supporting stress improvement and productivity improvement through chat analysis according to an embodiment of this specification. This embodiment effectively improves the stress and productivity of a subject, and as a result, it is possible to revitalize an organization and reduce the rate of mental illness. In a chat system, multiple participants can post (send) messages in real time and read (receive) the posted messages. The technique described below can also be applied to a system or service other than chat, such as a bulletin board system, in which multiple participants post text messages to each other. It can also be used to improve participants' conditions other than stress or productivity.

[0016] The system shown in Fig. 1 is a computer system. The computer system includes a chat analysis device 1 and an employee terminal 3, which communicate with each other via a network 5. Fig. 1 shows one employee terminal 3 as an example, but the employee terminals of a plurality of employees who participate in chats are connected to the network 5.

[0017] The chat analysis device 1 includes an employee classification estimation unit 10, a stress / productivity score estimation unit 11, an estimation result integration processing unit 12, a suggested behavior creation unit 13, a suggested behavior interest / effect judgment unit 14, a model training unit 15, an employee classification estimation model 16, a score estimation model 17, a score improvement behavior estimation model 18, and an interaction data collection unit 19. The chat analysis device 1 further stores a training data DB (database) 21, an interaction group DB 23, an improvement behavior DB 25, and an interaction data DB 27.

[0018] The employee classification estimation unit 10 estimates the personality type of the employee based on the content of the chat using an employee classification estimation model 16. The stress / productivity score estimation unit 11 estimates the stress score and productivity score of the employee based on the content of the chat using a score estimation model 17. Note that other types of scores may be used to improve other conditions of the employee.

[0019] The estimation result integration processing unit 12 generates data to be input to the score improvement behavior estimation model 18. The suggested behavior creation unit 13 decides the behavior to be suggested to the employee based on the stress score and the productivity score. The suggested behavior interest / effectiveness determination unit 14 determines the interest and effectiveness of the suggestion based on the response to the behavior suggested to the employee.

[0020] The model training unit 15 trains an employee classification estimation model 16, a score improvement action estimation model 18, and a score estimation model 17. The employee classification estimation model 16 estimates the personality type of an employee based on the employee's chat data. The score estimation model 17 calculates the employee's productivity score and stress score based on the employee's chat data and the employee classification.

[0021] The score improvement behavior estimation model 18 estimates a suggested behavior that can contribute to improving the employee's score based on the productivity score and the stress score. The interaction data collection unit 19 collects information on interaction groups and chat data for each interaction group from the employee terminal 3.

[0022] The employee terminal 3 includes a chat input / sending unit 31 and a chat display unit 32. The chat input / sending unit 31 accepts chat input by an employee and sends it to a server (not shown) that provides chat services. The chat display unit 32 displays a screen of chats by the employee and other employees and a screen presenting improvement actions proposed by the chat analysis device 1. The improvement actions can be displayed on a screen of chats with other employees, a screen of chats with the chat analysis device 1, or a screen of another application.

[0023] The functional units 10 to 18 in the chat analysis device 1 and the functional units 31 and 32 in the employee terminal 3 may be realized by a processor that executes a program, or may be realized by a hardware module implemented in accordance with each functional unit.

[0024] 2 is a diagram showing an example of a hardware configuration of the chat analysis device 1 according to an embodiment of the present specification. In the following, an example of the hardware configuration of the chat analysis device 1 will be described, but the employee terminal may also have a similar configuration.

[0025] The chat analysis device 1 includes a CPU (processor) 201 that executes various programs, a memory (main storage device) 202 that stores the various programs, and an auxiliary storage device 203 that stores various data. The CPU 201 can include one or more cores, and the memory 202 is, for example, a RAM that includes a volatile storage area. The auxiliary storage device 203 is, for example, a HDD (Hard Disk Drive) or a flash memory, and can provide a non-volatile storage area.

[0026] The chat analysis device 1 further includes an output device 204 for presenting information to a user, an input device 205 for inputting instructions, images, etc. from the user, and a communication device 206 for communicating with other devices. These are connected to each other by a bus 207.

[0027] The functional units of the chat analysis device 1 shown in Fig. 1 can be implemented, for example, by the CPU 201 operating according to a program. The CPU 201 reads and executes various programs from the memory 202 as necessary. The memory 202 stores the programs. In Fig. 2, the same names and symbols as those of the functional units shown in Fig. 1 are given to the programs.

[0028] Specifically, the memory 202 stores each program of an employee classification estimation unit 10, a stress / productivity score estimation unit 11, an estimation result integration processing unit 12, a suggested behavior creation unit 13, a suggested behavior interest / effect determination unit 14, a model training unit 15, an employee classification estimation model 16, a score estimation model 17, a score improvement behavior estimation model 18, and an interaction data collection unit 19. Each program is loaded into the memory 202 from the auxiliary storage device 203, for example, and executed by the CPU 201. Note that at least a part of the functional units may be configured with logic circuits.

[0029] The auxiliary storage device 203 stores data to be referred to or managed by various programs. In FIG. 2, the auxiliary storage device 203 stores a training data DB 21, an interaction group DB 23, an improvement action DB 25, and an interaction data DB 27.

[0030] The output device 204 is composed of devices such as a display, a printer, a speaker, etc. The input device 205 is composed of devices such as a keyboard, a mouse, a microphone, etc. The output device 204 presents the results of input from the user and also presents the results of processing by the chat analysis device 1. Instructions from the user are input to the chat analysis device 1 by the input device 205.

[0031] The communication device 206 receives data transmitted from other devices connected via a network, including the employee terminal 3, and transmits the results of processing by the chat analysis device 1 to the other devices. Note that some of the devices may be omitted. <Model training>

[0032] 3 shows an example of the configuration of training data 310 for the employee classification estimation model 16 and the score estimation model 17. The training data 310 is included in the training data DB 21. The employee classification estimation model 16 estimates the personality type of an employee from information obtained from a chat in which the employee participates. The score estimation model 17 calculates the stress score and productivity score of the employee from information obtained from a chat in which the employee participates.

[0033] 3, the training data 310 indicates explanatory variables input to each of the employee classification estimation model 16 and the score estimation model 17, and a target variable output to the employee classification estimation model 16 or the score estimation model 17. The explanatory variables are feature quantities input to the models.

[0034] Specifically, the training data 310 includes an employee ID field 311 and a period field 312. The explanatory variable fields include a number of posts field 313, a post sentiment analysis field 314, a number of predictive words field 315, an average post time field 316, and a number of post responses field 317. The objective variable fields include an employee classification field 318, a stress score field 319, and a productivity score field 320.

[0035] Each record in the training data 310 represents information about one employee in a chat in a particular social group for a particular period of time. A social group is a chat group. Multiple participants can post messages in the chat space of the social group at will, and all participants can see the posted messages.

[0036] When the explanatory variables of one record are input to the employee classification estimation model 16, the employee classification estimation model 16 outputs an estimated employee classification. When the explanatory variables of one record are input to the score estimation model 17, the score estimation model 17 outputs estimated values ​​of the stress score and the productivity score.

[0037] The employee ID column 311 indicates the employee ID. The period column 312 indicates the period of the chat for each record. The number of posts column 313 indicates the number of posts made in the chat by the employee indicated in the employee ID column 311 during the period indicated in the period column 312. The post emotion analysis column 314 indicates the emotion of the employee obtained from an analysis of the content of the posts. In the example of FIG. 3, emotions are represented by "+" indicating positive or "-" indicating negative. Note that it is possible to estimate emotions from text using, for example, a machine learning model based on deep learning. Emotion recognition technology is widely known, and details will be omitted.

[0038] The number of predictive words field 315 indicates the number of predictive words appearing in a post. Predictive words are, for example, predictive words for mental illness, and words that are the same as or similar to pre-set words are extracted. The average posting time field 316 indicates the average posting time of messages. The number of post responses field 317 indicates the number of responses to messages posted by others. For example, a "like" reaction corresponds to a response.

[0039] The number of posts, average time spent posting, and number of responses to posts indicate the posting behavior of employees. The results of the post sentiment analysis and the number of predictive words are values ​​based on the content of employees' posts. This information allows for a more appropriate estimation of the employee's condition. While it is possible to estimate the employee's condition using only one type of information, using two types of information allows for a more appropriate estimation.

[0040] The employee classification column 318 indicates the personality type of the employee. The stress score column 319 indicates the employee's stress score calculated from the explanatory variables of the corresponding record. A lower stress score indicates less stress. The productivity score column 320 indicates the employee's productivity score calculated from the explanatory variables of the corresponding record. A higher productivity score indicates higher productivity.

[0041] The model training unit 15 uses the training data 310 to train the employee classification estimation model 16 and the score estimation model 17. For example, these models 16 and 18 are configured as deep neural networks, and the model training unit 15 optimizes the parameters of the models 16 and 18 by backpropagation.

[0042] Note that the explanatory variables shown in FIG. 3 are examples, and some of these may be omitted or other explanatory variables may be added. For example, the amount of variation from the average posting time at the time of previous compilation may be added or may be replaced with the average posting time. Alternatively, the amount of variation in the number of responses to a post may be added or may be replaced with the number of responses to a post. The employee classification estimation model 16 and the score estimation model 17 may be rule-based models different from machine learning models.

[0043] As described above, the inputs of the classification model and the score calculation model can include variables based on the content of chat posts by employees, posting frequency, text volume, emotion analysis results, responses to chat suggested actions, and the amount of variation thereof. These explanatory variables represent changes in employees' moods and proactiveness in posting to chat, and can contribute to more accurate estimation of stress scores and productivity scores. Note that at least some of the above examples of input explanatory variables for the two models may be omitted, and other types of variables may be added.

[0044] 4 shows an example of the configuration of training data 350 for the score-improvement behavior estimation model 18. The training data 350 is included in the training data DB 21. The score-improvement behavior estimation model 18 estimates behavior that can improve an employee's stress score and productivity score based on the employee classification and the stress score and productivity score. By referring to both the classification and the score, more appropriate estimation is possible. Some of these may be omitted.

[0045] 4, the training data 350 indicates explanatory variables input to the score-improvement behavior estimation model 18 and objective variables output to the score-improvement behavior estimation model 18. Specifically, the training data 350 includes an employee ID column 351. The explanatory variable column includes a classification column 352, a productivity score at the time of suggested behavior column 353, a stress score at the time of suggested behavior column 354, a suggested behavior column 355, a number of people in a group column 356, a chat poster ratio column 357, and an average number of chats / day column 358. The objective variable column includes a productivity score change column 359 and a stress score change column 360.

[0046] Each record of the training data 350 indicates information about one employee in chats of one particular social group during a particular period of time. When the explanatory variables of one record are input to the score improvement behavior estimation model 18, the score improvement behavior estimation model 18 outputs estimated values ​​of the score stress score variance and the productivity score variance.

[0047] The employee ID column 351 indicates the employee ID. The classification column 352 indicates the employee's personality type. The productivity score at time of suggested action column 353 indicates the employee's productivity score when the score-improvement behavior estimation model 18 estimates an improvement behavior. The stress score at time of suggested action column 354 indicates the employee's stress score when the score-improvement behavior estimation model 18 estimates an improvement behavior.

[0048] The suggested action column 355 indicates an improvement action suggested by the score improvement action estimation model 18. In this example, it indicates a category of the suggested improvement action. For example, "walk" indicates making a suggestion about walking, and "sleep" means making a suggestion about sleep. "Information exchange group suggestion" means making a suggestion to join another exchange group.

[0049] The category of the improvement action to be proposed or the identifier of the interaction group in which participation is proposed is input to the score improvement action prediction model 18. By including participation in other interaction groups as candidates for the proposed action, the possibility of improving the employee's score can be increased.

[0050] The number of people in a group field 356 indicates the number of people belonging to an interaction group. The chat poster ratio field 357 indicates the ratio of the number of participants who posted a message within a specified period. Here, a participant who posted even once is counted as a poster. Note that only participants who posted more than a specified number of posts may be counted. The average number of chats / day field 358 indicates the average number of posts per day in an interaction group within a specified period. The unit time for the average number may be a length other than one day. Note that the specified periods for calculating values ​​in different fields may be the same or different.

[0051] The productivity score change field 359 indicates the amount of change in the productivity score from the productivity score at the time of the action suggestion, calculated from the explanatory variables of the corresponding record. The stress score change field 360 indicates the amount of change in the stress score from the stress score at the time of the action suggestion, calculated from the explanatory variables of the corresponding record. If the amount of change in the stress score is negative, it indicates that the mental state of the employee is improving.

[0052] By inputting data on different proposed actions into the score improvement action estimation model 18, it is possible to obtain score change values ​​for the different proposed actions. By selecting actions that favorably change the score and proposing them to employees, it is possible to expect improvements in employee stress and productivity.

[0053] The model training unit 15 uses the training data 350 to train the score improvement behavior estimation model 18. For example, the score improvement behavior estimation model 18 is configured with a deep neural network, and the model training unit 15 optimizes parameters of the score improvement behavior estimation model 18 by the backpropagation method.

[0054] 4 are examples, and some of them may be omitted or other explanatory variables may be added. The score-improvement behavior estimation model 18 may be a rule-based model different from a machine learning model.

[0055] The input of the score improvement behavior prediction model 18 includes information on the interaction group in addition to the employee's score, which makes it possible to more accurately predict an appropriate interaction group for the employee. <Action suggestion>

[0056] The following describes an example of a process of suggesting actions to employees by the chat analysis device 1. The chat analysis device 1 stores an employee classification estimation model 16, a score improvement action estimation model 18, and a score estimation model 17 trained by a model training unit 15 using training data 310 and 350 in a training data DB 21. The chat analysis device 1 analyzes chats of employees and suggests actions to the employees to improve their stress scores and / or productivity scores.

[0057] The interaction data collection unit 19 of the chat analysis device 1 collects chat data in each employee's interaction group in order to propose actions. The interaction data collection unit 19 analyzes the collected data and updates the interaction group DB 23 and the interaction data DB 27 according to the analysis results.

[0058] FIG. 5 shows an example of the configuration of the social group DB 23. The social group DB 23 manages information about social groups. A plurality of employees belong to a social group and participate in chats in the social group. An employee chats with other employees in the social group to which the employee belongs. For each social group, a purpose and an action (improvement action) for the purpose are defined. Participants in the social group post about the improvement action in the chat. Each employee belongs to one or more social groups.

[0059] For example, the improvement behavior and goal of one exchange group is "improving stress by making morning walking a habit." The improvement behavior and goal of another exchange group is "improving cognitive function and productivity by fast walking." The improvement behavior and goal of another exchange group is "preventing mental illness with quality sleep."

[0060] The interaction data collection unit 19 periodically collects, for example, interaction data between employees in interaction groups in which each employee participates from multiple employee terminals 3, analyzes the collected interaction data, and updates the interaction group DB 23 based on the analysis results. Chat data may be collected from a system that provides a chat service. The collected interaction data includes information on each interaction group itself and chat data for each interaction group.

[0061] 5, the communication group DB 23 includes an improvement action field 231, a purpose field 232, a group member field 233, a chat message submission number ratio field 234, and an average number of chats per day field 235.

[0062] The interaction group DB 23 further includes a column showing the number of employees in each employee classification in the interaction group. In Fig. 5, a column 236 showing the number of employees in classification (A) and a column 237 showing the number of employees in classification (B) are shown as examples. The interaction data collection unit 19 can determine the number of employees in each classification from the estimation results of the employee classification estimation model 16 based on the past interaction data.

[0063] The improvement action column 231 indicates the action that the participants of the communication group talk about in the chat. The improvement action and purpose are defined in advance for each communication group. For example, if the improvement action is "walking", the participants post messages in the chat space reporting the results of their own walking, useful information related to walking, responses to other people's posts, etc. The purpose column 232 indicates the purpose of the communication group. The number of people in group column 233 indicates the number of people belonging to the group.

[0064] The chat poster ratio field 357 indicates the ratio of the number of participants who posted messages within a specified period. Here, any participant who posted at least once is counted as a poster. It is also possible to count only participants who posted more than a specified number of posts. The average number of chats / day field 358 indicates the average number of posts per day in the interaction group within a specified period. The average unit of time may be different from one day.

[0065] The group number column 233, the chat poster percentage column 234, and the average number of chats per day column 235 respectively store the same type of data as the group number column 356, the chat poster percentage column 357, and the average number of chats per day column 358 in the training data 350 of the score improvement behavior estimation model 18.

[0066] The data in the improvement action column 231 and the objective column 232 are fixed for each interaction group. The data in the other columns 233 to 237 are updated as needed according to the analysis results of the interaction data collection unit 19.

[0067] 6 shows an example of the configuration of the interaction data DB 27. The interaction data DB 27 stores features to be input to the employee classification estimation model 16 and the score estimation model 17. The interaction data collection unit 19 periodically collects, for example, interaction data between employees in interaction groups in which each employee participates, from multiple employee terminals 3, analyzes the collected interaction data to extract predetermined features, and stores information on the analysis results in the interaction data DB 27. The interaction data collection unit 19 stores data indicating the contents of the chat in the interaction data DB 27 or another database not shown.

[0068] The interaction data DB27 includes items similar to the explanatory variables of the training data 310 of the employee classification estimation model and the score estimation model shown in Fig. 3. Each record indicates information about posts of one employee in one interaction group during a specific period.

[0069] Specifically, the interaction data DB 27 includes an employee ID column 271 and a period column 272. It further includes a number of posts column 273, a post sentiment analysis column 274, a number of predictive words column 275, an average posting time column 276, a number of post responses column 277, and an interaction group column 278. These columns store the same types of data as the columns with the same names in the training data 310 for the employee classification estimation model and the score estimation model. The interaction group column 278 indicates the interaction group of the chat.

[0070] In the example of FIG. 6, an employee with ID1 belongs to social group A and social group B, and information about posts by the employee in each social group is periodically collected and stored. An employee with ID2 belongs to social group A, and information about posts by the employee in the group is periodically collected and stored. Old records may be deleted in a timely manner.

[0071] Fig. 7 shows a flowchart of a processing example of the chat analysis device 1. In step S11, the interaction data collection unit 19 collects employee interaction data from each employee terminal 3. As described with reference to Figs. 5 and 6, the interaction data collection unit 19 updates the interaction group DB 23 and the interaction data DB 27 based on the collected interaction data. The update frequency of the interaction group DB 23 and the interaction data DB 27 may be different.

[0072] The process is described below for one employee in one social group. The process described below can be performed for each employee in each social group.

[0073] In step S12, the employee classification estimation unit 10 estimates the type of the target employee of the target interaction group based on the data stored in the interaction data DB 27, using the employee classification estimation model 16. The employee classification estimation unit 10 inputs the latest data of the target employee of the target interaction group into the employee classification estimation model 16 to determine the type of the target employee.

[0074] Furthermore, in step S13, the stress / productivity score estimation unit 11 estimates the type of the target employee in the target interaction group based on the data stored in the interaction data DB 27, using the score estimation model 17. The stress / productivity score estimation unit 11 inputs the data of the target employee in the target interaction group into the score estimation model 17, and calculates the stress score and productivity score of the target employee. For example, the productivity score and stress score are calculated based on the interaction data for a predetermined period immediately before the improvement action is estimated.

[0075] Next, in step S14, the chat analysis device 1 calculates estimated fluctuations of the stress score and the productivity score for each improvement behavior based on the employee type, the stress score, and the productivity score.

[0076] First, the estimation result integration processing unit 12 generates data to be input to the score-improvement behavior estimation model 18. Fig. 8 shows an example of input data 400 to the score-improvement behavior estimation model 18. Items of the input data set shown in Fig. 8 are the same as the items of the explanatory variables of the training data 350 of the score-improvement behavior estimation model shown in Fig. 4.

[0077] Specifically, the input data 400 is composed of a classification field 402, a productivity score at the time of suggested action field 403, a stress score at the time of suggested action field 404, a suggested action field 405, a number of people in a group field 406, a percentage of the number of people posting in chat messages field 407, and an average number of chat messages per day field 408. The data periods for calculating values ​​in different fields may be the same or different.

[0078] Each record indicates one input data set. A plurality of input data sets are sequentially input to the score improvement behavior estimation model 18, and a stress score variation amount and a productivity score variation amount are output. All records of the input data shown in FIG. 8 indicate combinations of information of a target employee in a target interaction group and different suggested behaviors.

[0079] The estimation result integration processing unit 12 stores the type of the target employee estimated in step S12 in all cells of the classification column 402. The estimation result integration processing unit 12 acquires the productivity score and stress score estimated in step S13, and stores them in all cells of the productivity score at time of suggested action column 403 and the stress score at time of suggested action column 404. The estimation result integration processing unit 12 stores different actions that are set in advance in cells of the suggested action column 405.

[0080] Furthermore, the estimation result integration processing unit 12 acquires the values ​​of the number of people in a group, the ratio of the number of people posting in chat messages, and the average number of chat messages / day of the target interaction group from the interaction group DB 23, from the number of people in a group field 233, the ratio of the number of people posting in chat messages field 234, and the average number of chat messages / day field 235. The estimation result integration processing unit 12 stores the acquired values ​​in all cells of the number of people in a group field 406, the ratio of the number of people posting in chat messages field 407, and the average number of chat messages / day field 408.

[0081] Furthermore, the suggested behavior creation unit 13 sequentially inputs the input data sets (records) of the input data 401 to the score improvement behavior estimation model 18, and obtains estimated values ​​of the fluctuation amount of the stress score and estimated values ​​of the fluctuation amount of the productivity score for each input data set.

[0082] In one embodiment of the present specification, the suggested action creation unit 13 may correct the stress score and the productivity score based on the employee's reaction to the suggested actions of a predetermined number of times in the past (for example, the previous time). This makes it possible to re-suggest more appropriate action candidates. Note that the employee's reaction to the previous suggested action does not need to be referenced.

[0083] The proposed behavior interest / effect judgment unit 14 searches for employee responses to the previous proposal in the chat data collected by the interaction data collection unit 19. The responses can be obtained from "Like" posts to the proposed message and the contents of messages posted after the proposal. Note that the method of text analysis is widely known, so details will be omitted.

[0084] For example, when a "Like" is posted or a related message is posted, the proposed behavior interest / effectiveness determination unit 14 may determine that there was interest or effectiveness, and improve the stress score and productivity score of the same type of behavior as the previous time by a predetermined value. When there is no response, or when a message denying the proposed behavior is posted, the proposed behavior interest / effectiveness determination unit 14 may determine that there was no interest or effectiveness, and may worsen the stress score and productivity score of the same type of behavior as the previous time by a predetermined value. The correction value may be changed depending on the number of responses or the number of related messages posted. When information on the actual behavior of the employee is available, the presence or absence of behavior in accordance with the proposal may be referenced.

[0085] Next, in step S15, the chat analysis device 1 executes the processes of steps S12 to S14 for each of the other participants in the target interaction group. As a result, an estimated value of the variation amount of the stress score and an estimated value of the variation amount of the productivity score for each of the other participants in the target interaction group are obtained.

[0086] The proposed action creation unit 13 stores the estimated values ​​of the fluctuation amount of the stress score and the estimated values ​​of the fluctuation amount of the productivity score of all participants in the target social group in the improvement score table. Fig. 9 shows an example of the configuration of the improvement score table 450. In the example of Fig. 9, the employee with ID1 is the target employee, and employees with ID2 and ID3 are other participants in the target social group. Candidates for the proposed improvement action are walking, sleeping, eating, joining social group A, and joining social group B.

[0087] In the next step S16 and thereafter, the proposed action creation unit 13 determines the (type of) improvement action to be proposed and the range of employees to which the improvement action is to be proposed. In step S16, the proposed action creation unit 13 determines the improvement action to be proposed to the target employee based on the two score variation values ​​of the target employee. For example, if the two score variation values ​​of the improvement action exceed their respective thresholds, it is determined that the improvement action is to be presented to the target employee. Alternatively, it may be determined based on the sum of the two score variation values ​​or a value input into a specified function. There may be one or more improvement actions presented. For example, an improvement action that is estimated to be the best from the two score variation values ​​may be selected. Note that only one of the score variation values ​​may be referenced.

[0088] Next, the proposed action creation unit 13 estimates the range of effect of the improvement action selected for the target employee based on the statistical value of the score fluctuation estimate value of the improvement action. For example, the proposed action creation unit 13 calculates the average value and variance value of the fluctuation value of the stress score and the productivity score within the interaction group. The average value is an index indicating the improvement effect of the entire interaction group, and the variance value is an index indicating the variation of the improvement effect within the interaction group.

[0089] The average value may be a simple average value (with a common weighting coefficient of 1) or a weighted average value, and the median value may be used instead of the average value. The standard deviation or the maximum value of the difference between the fluctuation values ​​may be used instead of the variance value.

[0090] The proposed action creation unit 13 compares the average value of the fluctuation value of the stress score and the productivity score with the respective thresholds, and further compares the variance value of the fluctuation value of the stress score and the productivity score with the respective thresholds. If the average value of the fluctuation value of the two scores exceeds the threshold and the variance value is less than the threshold (S16: YES), it is estimated that there is an effect on the entire interaction group. If not (S16: NO), it is determined that the effect on the entire interaction group is small.

[0091] If the improvement action is effective for the entire communication group (S16: YES), in step S18, the proposed action creation unit 13 presents the improvement action in the group chat (S18). If the improvement action is not effective for the entire communication group (S16: YES), in step S19, the proposed action creation unit 13 presents the improvement action only to the target employee. After that, the flow returns to step S13, and the score is re-estimated, the score variation value is re-estimated, and the improvement action is re-suggested.

[0092] In one embodiment of the present specification, the proposed action creation unit 13 selects a sentence to be presented to the employee from the improvement action DB 25. For example, the proposed action creation unit 13 selects a sentence of a message from the improvement action DB 25 based on the type of improvement action decided to be proposed and the contents of the message in the chat. The method of creating the message to be presented is known as a so-called chatbot technology, and the details will be omitted here.

[0093] In the above example, the score estimation model 17 is used to estimate the score of the type of improvement action. In another example, an improvement action indicating more specific content may be input to the score estimation model 17. The input improvement action may be a template indicating more specific content, or may be a message that is actually presented to the employee. When a template is input, a message to be presented may be selected from the improvement action DB 25 based on the template and the actual chat content. In addition, the suggested action creation unit 13 may process the message stored in the improvement action DB 25 according to the chat content.

[0094] An example of a chat screen presented to an employee is described below. Figures 10A to 10D show examples of screens displayed on the employee terminals 3 of all employees or only one employee in one interaction group. The improvement action of the interaction group in this example is "walk."

[0095] Fig. 10A shows an example of messages in a chat room. Fig. 10A shows messages between two people as an example. As described above, the chat analysis device 1 analyzes chat exchange data as shown in Fig. 10A, and estimates the effects of improvement actions on productivity and stress.

[0096] Fig. 10B shows a display example of an improvement action suggested by the chat analysis device 1. In the example of Fig. 10B, the chat analysis device 1 posts a message 505 indicating an improvement action in a group chat. The improvement action estimated to be effective for the entire interaction group can be shown to all participants in the interaction group.

[0097] FIG. 10C shows another display example of improvement actions proposed by the chat analysis device 1. In the example of FIG. 10C, the chat analysis device 1 presents improvement actions on a proposed action presentation screen dedicated to one participant. This makes it possible to present improvement actions only to employees who need them. Participants can return a "Like" response 511 to the proposed action, and also input their actual number of steps 512.

[0098] 10D shows an example of responses of participants to an improvement action proposed in a chat room by the chat analysis device 1. A "Like" 521 and a related message 522 are returned in response to an improvement action suggestion message 505. As described above, the chat analysis device 1 can correct the score variation value based on the responses 521, 522 from participants to the proposed improvement action. Similarly, the chat analysis device 1 can correct the score variation value based on the "Like" response 511 and the change in the actual number of steps 512 (change in actual action in response to the action suggestion) shown in FIG. 10C.

[0099] The present invention is not limited to the above-mentioned embodiment, but includes various modified examples. For example, the above-mentioned embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to those including all of the configurations described. It is also possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace a part of the configuration of each embodiment with another configuration. In addition, the above-mentioned configurations, functions, processing units, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. In addition, the above-mentioned configurations, functions, etc. may be realized in software by a processor interpreting and executing a program that realizes each function. Information such as the program, table, file, etc. that realizes each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card or SD card. In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are connected to each other. [Explanation of symbols]

[0100] 1. Chat analysis device 3. Employee terminals 5. Network 10 Employee Classification Estimation Department 11 Stress and Productivity Score Estimation Section 12 Estimation result integration processing unit 13 Proposal Action Creation Department 14 Proposal Action Interest and Effect Assessment Section 15 Model Training Department 16 Employee Classification Estimation Model 17 Score Estimation Model 18 Score improvement behavior estimation model 19 AC Data Collection Division 21 Training Data DB 23 Exchange Group DB 25 Improvement action DB 27 Exchange Data DB 201 CPU 202 Memory 203 Auxiliary storage device 206 Communication Devices

Claims

1. A device for suggesting actions to participants in a social group, comprising: A processor; A storage device for storing a program executed by the processor; Including, The processor, Extracting a feature of the first participant from exchange data including text messages posted by multiple participants of the first exchange group; Calculating a predetermined score for the first participant based on at least a part of the feature amount of the first participant; Based on the predetermined score, a change value of the predetermined score due to each candidate improvement action that improves the predetermined score is estimated; the candidate remedial action includes participating in a social group different from the first social group; The apparatus determines a suggested improvement action for the first participant based on the change in the predetermined score.

2. 2. The apparatus of claim 1, The processor, recalculating the predetermined score of the first participant after obtaining information of the first participant's response to the proposed improvement action; re-estimating a change in the predetermined score due to each candidate improvement action that improves the predetermined score based on the recalculated predetermined score and the response information; The apparatus determines a re-suggested improvement action to the first participant based on the variation of the re-estimated predetermined score.

3. 2. The apparatus of claim 1, The processor, extracting the feature amount of each of the plurality of participants from the interaction data; estimating a variation value of the predetermined score for each of the plurality of participants based on at least a part of the feature amount of each of the plurality of participants; The apparatus determines whether to present the proposed improvement action to the first interaction group based on a variation value of the predetermined score for each of the multiple participants.

4. 2. The apparatus of claim 1, The processor, estimating a type of the first participant based on at least a portion of the characteristic amount of the first participant; The apparatus estimates a change in the predetermined score due to each of the candidate remedial actions that improve the predetermined score based on the type.

5. 5. An apparatus according to claim 1 or 4, comprising: The processor, The apparatus estimates a change in the predetermined score due to each of candidate improvement actions that improve the predetermined score based on information of the first interaction group.

6. 2. The apparatus of claim 1, The feature amount includes a feature amount indicating a manner in which the message was posted and a feature amount extracted from the message content.

7. 2. The apparatus of claim 1, The predetermined score is one of a stress score or a productivity score.

8. 8. The apparatus of claim 7, The processor calculates a plurality of scores including the predetermined score; the plurality of scores includes the stress score and the productivity score, The processor determines a suggested improvement action for the first participant based on the variation of the plurality of scores.

9. 1. A method of suggesting actions to participants in an interaction group, comprising: The device extracts a feature of a first participant from interaction data including text messages posted by multiple participants of a first interaction group; The device calculates a predetermined score of the first participant based on at least a part of the feature amount of the first participant; The device estimates a change in the predetermined score due to each candidate improvement action that improves the predetermined score based on the predetermined score; the candidate remedial action includes participating in a social group different from the first social group; The method, wherein the device determines a suggested improvement action for the first participant based on a change in the predetermined score.

Images