Information processing device, information processing method, and program

Abstract

An information processing device, an information processing method, and a program capable of suitably acquiring meeting characteristics that indicate the tendency of a target person to participate in a meeting. An information processing device includes an acquisition unit that acquires raw information that represents the contents of a meeting and includes a plurality of pieces of information, and a meeting feature extraction unit that extracts meeting features from the plurality of pieces of information in the raw information.

Description

[Technical field]

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

[0002] Conventionally, an invention has been disclosed in which conference information including audio data of speech by speakers during a conference is acquired, and the contribution of each speaker is calculated based on the information converted into text (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2024-118317 A Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional techniques are limited to deriving the presence of a speaker in a single conference, and may not be able to appropriately acquire conference features that indicate the tendency of a target person to participate in the conference.

[0005] The present invention has been made in consideration of the above circumstances, and one of its objectives is to provide an information processing device, an information processing method, and a program that are capable of preferably acquiring conference characteristics that indicate the tendency of a target person to participate in a conference. [Means for solving the problem]

[0006] One aspect of the present invention is an information processing device comprising an acquisition unit that acquires raw information representing the contents of a meeting and including multiple item information, and a meeting feature extraction unit that extracts meeting features from the multiple item information in the raw information. Effect of the Invention

[0007] According to one aspect of the present invention, it is possible to preferably acquire meeting characteristics that indicate the tendency of a subject person to participate in a meeting. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 illustrates an example of a usage environment of an information processing device. [Diagram 2] 1 illustrates an example of a configuration of an information processing device 100. [Diagram 3] 13 is a flowchart showing an example of the overall processing flow of the information processing device 100, focusing on the determination unit 120. [Figure 4] 13 is a flowchart (part 1) showing an example of the flow of processing by a first determination unit 122. [Diagram 5] 13 is a flowchart (part 2) showing an example of the flow of processing by the first determination unit 122. [Figure 6] 13 is a diagram conceptually showing the process of a second determination unit 124. FIG. [Figure 7] 11 is a flowchart showing an example of the overall processing flow of the information processing device 100, focusing mainly on the conference feature extraction section 130 and the index value derivation section 140. [Figure 8] FIG. 13 is a diagram for explaining centrality. [Figure 9] FIG. 11 is a diagram showing a first example of an image showing an extraction result of a meeting feature. [Figure 10] FIG. 11 is a diagram showing a second example of an image showing an extraction result of meeting features. [Figure 11] FIG. 13 is a diagram showing a third example of an image showing an extraction result of meeting features. [Figure 12] FIG. 13 is a diagram showing an example of a screen showing a derivation result of an index value. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] [overview] Hereinafter, with reference to the drawings, an embodiment of an information processing device, an information processing method, and a program according to the present invention will be described. The information processing device is a device that acquires original information in which a plurality of item information is associated with one or more schedules of a managed person, and based on the original information, determines whether each schedule is a meeting schedule, extracts meeting characteristics from the original information, and derives an index value that a certain managed person should aim for. The managed person is, for example, an employee of a certain company, and the processing results of the information processing device are used for proposals to improve the working environment of employees, personnel evaluation, etc. Below, the functions of the information processing device will be described based on this premise.

[0010] [Usage environment example] 1 is a diagram showing an example of a usage environment of an information processing device 100. The information processing device 100 acquires raw information from a calendar management system 50 accessed by a terminal device 10 to be managed, for example.

[0011] The managed person terminal device 10 is, for example, a company personal computer (which may include a tablet terminal or a smartphone) used by a company employee, and includes at least a processor such as a CPU (Central Processing Unit), a communication device, a display device, an input device, and a storage device for storing programs and data. A calendar application 20 runs on the managed person terminal device 10. The calendar application 20 may be a standalone application program, or an application program associated with an e-mail application or an online calling application. The calendar application 20 communicates with the calendar management system 50 via a network NW. The network NW includes some or all of the Internet, a WAN (Wide Area Network), a LAN (Local Area Network), a cellular network, a wireless base station, a provider device, and the like.

[0012] The calendar management system 50 includes, for example, a calendar processing unit 52 and a storage unit 60. The calendar processing unit 52 edits calendar information 62 in response to an operation performed by an employee on the calendar application 20. In addition, in response to a viewing request from the calendar application 20, the calendar processing unit 52 enables viewing of the schedules of the employee and other employees who have agreed to viewing. The calendar information 62 is information that includes source information, which will be described later. The calendar management system 50 is commonly known under a name such as "XX Calendar," and a detailed description thereof will be omitted in this specification.

[0013] The information processing device 100 is operated by, for example, an administrator terminal device 200. The administrator terminal device 200 is, for example, a terminal device used by a human resources officer in a company, and may be similar in terms of hardware to the managed person terminal device 10. Alternatively, the information processing device 100 may be configured by installing a program for operating as the information processing device 100 in the administrator terminal device 200. In other words, the information processing device 100 and the administrator terminal device 200 may be configured as one unit.

[0014] [composition] FIG. 2 is a diagram showing an example of the configuration of the information processing device 100. The information processing device 100 includes, for example, an acquisition unit 110, a determination unit 120, a conference feature extraction unit 130, an index value derivation unit 140, an output unit 150, and a storage unit 170. The determination unit 120 includes, for example, a first determination unit 122 and a second determination unit 124. The components other than the storage unit 170 are realized by, for example, a hardware processor such as a CPU executing a program (software). Some or all of these components may be implemented using LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable The program may be realized by hardware (including circuitry) such as a Gate Array (GPU) or a Graphics Processing Unit (GPU), or may be realized by a combination of software and hardware. The program may be stored in advance in a storage device (a storage device having a non-transient storage medium) such as a Hard Disk Drive (HDD) or a flash memory, or may be stored in a removable storage medium (non-transient storage medium) such as a DVD or CD-ROM, and installed in the storage device by mounting the storage medium in a drive device.

[0015] The storage unit 170 is a HDD, a flash memory, a RAM (Random Access Memory), etc. The storage unit 170 may be a NAS (Network Attached Storage) device that the information processing device 100 can access via a network. The storage unit 170 stores information such as determination rule information 172, a trained model 174, and an index value / comment conversion table 176.

[0016] [Meeting Judgment] FIG. 3 is a flowchart showing an example of the overall processing flow of the information processing device 100 with the determination unit 120 at the center. First, the acquisition unit 110 acquires raw information in which a plurality of item information items are associated with one or more plans of employees managed by the calendar management system 50 (S1). The acquisition unit 110 acquires the raw information by using, for example, an API (Application Programming Interface) provided by the calendar management system 50. The raw information is, for example, information items associated with each plan, such as the employee who made the plan, a summary text about the plan, an explanation text, an online conference flag, a venue, a publication status, a start time and an end time, a duration, a number of people who share the plan, a response status, a predetermined number of people who have not responded, and a frequency. The online conference flag is a flag that is set to 1 when the URL (Uniform Resource Locator) of the online conference is included in the summary text or explanation text, or when characters indicating an online conference are included in the summary text or explanation text, and is set to 0 otherwise. This flag may be assigned by the calendar management system 50 or by the acquisition unit 110. In the case of an online conference, a URL is described in the venue. The number of shared employees is the number of employees with whom the schedule was shared, the response status is information indicating whether the employee responded that they would participate, that they would not participate, or that they had not responded, and the specified number of non-responders is the number of employees with whom the schedule was shared who have not responded that they would not participate.

[0017] Next, the judgment unit 120 performs a judgment by the first judgment unit 122 and a judgment by the second judgment unit 124 based on multiple item information, for example, for each schedule in the original information, and makes an integrated judgment of the judgment results to determine whether each schedule is a meeting schedule or not (S2 to S4).

[0018] The first determination unit 122 performs a determination based on a rule defined by the determination rule information 172 (S2). The determination rule information 172 may be edited by the administrator terminal device 200, and in that case, the contents of the determination by the first determination unit 122 may be changed similarly. Figures 4 and 5 are flowcharts showing an example of the flow of processing by the first determination unit 122.

[0019] First, the first determination unit 122 determines whether the publication status of the original information is "private" or not (S10). If the publication status of the original information is "private", the process proceeds to the processing of the flowchart in Fig. 5. If the publication status of the original information is not "private", that is, if it is "public", the first determination unit 122 determines whether the reply status is "not attending" or not (S11). If the reply status is "not attending", the first determination unit 122 determines that the schedule is not a meeting (S17).

[0020] If the answer status is not "not attending" (including "attending" and "no answer"), the first determination unit 122 determines whether the length of time is a predetermined time or more (e.g., about 8 hours) (S12). If the length of time is the predetermined time or more, the first determination unit 122 determines that the event is not a meeting (S17).

[0021] If the length of time is less than the predetermined time, the first determination unit 122 determines whether the online conference flag is zero and the venue is blank (S13). If the online conference flag is zero and the venue is blank, the first determination unit 122 determines that the event is not a conference (S17).

[0022] If the online conference flag is not zero or the venue is not blank, the first determination unit 122 determines whether the number of sharers and the predetermined number of non-responders are both less than 2 (S14). If the number of sharers and the predetermined number of non-responders are both less than 2, the first determination unit 122 determines that the event is not a conference (S17).

[0023] If at least one of the number of sharing users and the number of predetermined non-answering users is 2 or more, the first determination unit 122 determines whether the summary text contains a predetermined word (S15). The predetermined word is a word that indicates that the event is not a meeting, and is, for example, "no meeting", "no call time", "fire drill", "no overtime day", "no overtime day", "no overtime day", "no meeting setting", "mtg_ng", "leave early", "half day off", "morning off", "afternoon off", "morning pay", "afternoon pay", "hourly leave", "paid leave use", "paid leave use", "flextime recommended", "flextime attendance", "flextime attendance", "flextime acquisition", "flextime leaving", "flextime early leave", "flextime work", "flextime DAY", "flextime day", "work block", "safety confirmation", "(black brackets) movement", "(movement)", and "outside of work". If the summary text contains a predetermined word, the first determination unit 122 determines that the event is not a meeting (S17).

[0024] If negative determination results are obtained in all of S10 to S16, the first determination unit 122 determines that the schedule is a public conference (S16).

[0025] 5, when the disclosure status of the original information is "private", the first determination unit 122 determines whether the answer status is "not attending" (S21). When the answer status is "not attending", the first determination unit 122 determines that the schedule is not a meeting (S27).

[0026] If the answer status is not "not attending" (including "attending" and "no answer"), the first determination unit 122 determines whether the length of time is a predetermined time or more (e.g., about 8 hours) (S22). If the length of time is the predetermined time or more, the first determination unit 122 determines that the event is not a meeting (S27).

[0027] If the length of time is less than the predetermined time, the first determination unit 122 determines whether the online conference flag is zero and the venue is blank (S23). If the online conference flag is zero and the venue is blank, the first determination unit 122 determines that the event is not a conference (S27).

[0028] If the online conference flag is not zero or the venue is not blank, the first determination unit 122 determines whether the number of sharers is less than two (S24). If the number of sharers is less than two, the first determination unit 122 determines that the event is not a conference (S27). The process of S24 differs from the process of S14 in Fig. 5. In this way, the first determination unit 122 performs at least partially different determination processes for public events and private events.

[0029] If at least one of the number of sharing users and the predetermined number of non-answering users is equal to or greater than 2, the first determination unit 122 determines whether the summary contains a predetermined word (S25). The predetermined word may be the same as in S15. If the summary contains the predetermined word, the first determination unit 122 determines that the event is not a meeting (S27).

[0030] If negative determination results are obtained in all of S21 to S26, the first determination unit 122 determines that the schedule is a closed meeting (S26).

[0031] Returning to FIG. 3, the second determination unit 124 inputs the original information to the trained model 174 and determines whether the schedule is a meeting or not based on the obtained output (S3). FIG. 6 is a diagram conceptually illustrating the processing of the second determination unit 124. The trained model 174 is created by a method such as gradient boosting or backpropagation, for example, with the original information (or an edited version of the original information) as learning data, with teacher words (e.g., "vacation" or "day off", which simply indicate that the schedule is not a meeting) described in the summary or description of the original information as negative examples and other words as positive examples. The second determination unit 124 inputs the original information to the original information to obtain a score indicating the probability that the schedule is a meeting. Here, the second determination unit 124 may apply a threshold value to the score to determine whether the schedule is a meeting or not, or may use the score as the determination result as it is.

[0032] 3, the determination unit 120 combines the determination results of S2 and S3 to make a final determination as to whether the schedule is a meeting or not (S4). For example, when the determination result of S3 is expressed as a score, the determination unit 120 makes the final determination by various methods, such as determining that it is a meeting if the score is 0.8 or more, and that it is not a meeting if the score is less than 0.2, and adopting the determination result of S2 if the score is 0.2 or more and less than 0.8.

[0033] The output unit 150 outputs the determination result by the determination unit 120 to the administrator terminal device 200, and causes the administrator terminal device 200 to display a screen showing the determination result (S5). The output unit 150 may render an image by itself, or may output parameters for constructing an image to the administrator terminal device 200 to generate an image. The output unit 150 may have a function of a web server, for example, and cause the administrator terminal device 200 to display the determination result in response to a request from a browser running on the administrator terminal device 200. The same applies to the following explanation.

[0034] According to the above-described process of "meeting determination," it is possible to appropriately determine whether or not an event is a meeting event based on information about the event.

[0035] [Meeting feature extraction & index value derivation] The following describes the processing of the meeting feature extraction unit 130 and the index value derivation unit 140. Fig. 7 is a flowchart showing an example of the overall processing flow of the information processing device 100, focusing on the meeting feature extraction unit 130 and the index value derivation unit 140. First, similar to the processing of the flowchart in Fig. 3, the acquisition unit 110 acquires raw information in which multiple pieces of item information are associated with one or more plans of an employee managed by the calendar management system 50 (S30).

[0036] Next, the meeting feature extraction unit 130 organizes the raw information determined to be a meeting by the determination unit 120 (an example of "raw information that indicates the content of a meeting and includes multiple item information") for each employee to derive meeting features (S31). The meeting features include, for example, the meeting volume, number of meetings, and meeting time ratio that indicate the amount of communication of an individual during an observation period (e.g., several months), the unique volume that indicates the breadth of communication of an individual, the centrality that indicates the number and degree of intermediation of an individual's connections, and the connection volume that indicates the size of an individual's connections.

[0037] The amount of meetings is the cumulative time that the employee of interest (hereinafter simply referred to as the employee, an example of a "second employee") participated in. The number of meetings is the number of meetings that the employee participated in. The percentage of meeting time is the percentage of working hours, including overtime, that were spent in meetings.

[0038] The unique amount is the number of unique employees who participated in the meeting together during the observation period (not the total number of people, but if they participated multiple times, they are counted as one person, and excluding optional participants).

[0039] Centrality is a numerical value (or discretized to show it in stages) of graph information showing the relationship between employees who participated in the same conference. Examples of centrality include degree centrality, closeness centrality, betweenness centrality, and eigenvector centrality. FIG. 8 is a diagram for explaining centrality. In this graph, the nodes represent employees, the edges connecting the nodes represent the connections to the same conference, and the length of the edge represents the number of conferences they participated in together (the more the number, the shorter the edge). Alternatively, the length of the edge may be an index of the closeness of the department. The degree centrality is a numerical value that increases with the number of other employees connected, the closeness centrality is the number of other employees who are close to each other, the betweenness centrality is a numerical value that increases with the position of the group that mediates between different groups (for example, a group with a predetermined number or more employees connected), and the eigenvector centrality is the number of other employees with high degree centrality connected to the employee. As shown in the figure, employees A and D have high degree centrality, employee B has high closeness centrality, employee C has high betweenness centrality, and employee E has high eigenvector centrality. From these meeting characteristics, the following insights can be obtained. For example, employees A and D with high degree centrality and employee B with high closeness centrality are considered to be cohesive types, with close and closed characteristics, and strengths in trust, norms, and idea execution, but weaknesses in siloed or localized information and knowledge leading to narrow-minded decision-making. On the other hand, employee C with high betweenness centrality is considered to be an intermediary type, with sparse and open characteristics, and strengths in disseminating information and ideas and creating new ideas, but weaknesses in weak mutual trust and consensus building, and difficulty in conveying tacit knowledge and complex information.

[0040] The amount of connection is a value that represents the distribution of affiliations of managed persons who participated in the same conference. The amount of connection is expressed, for example, as a vector (R1, R2, R3, R4) in which the vector elements from the left are the proportion R1 of other employees who participated in the conference who are in a team with the target employee, the proportion R2 of those who are in a department, the proportion R3 of those who are in the headquarters, and the proportion R4 of those who are outside the headquarters. In general, the larger the vector element on the right, the higher the degree of the individual connection is considered to be.

[0041] The output unit 150 outputs the extracted results of the conference features to the manager terminal device 200, and causes the manager terminal device 200 to display a screen showing the extracted results of the conference features (S33). FIG. 9 is a diagram showing a first example of an image showing the extracted results of the conference features that the output unit 150 causes the manager terminal device 200 to display. In this image, the average number of meetings per month and the ratio of conference time are shown by general employee / leader. The labels C, B, A, and S in the legend indicate the employee evaluation ranks (described later). In this way, the output unit 150 may cause the manager terminal device 200 to display graphs or the like showing the average number of meetings per month and the ratio of conference time by general employee / leader and further by rank.

[0042] 10 is a diagram showing a second example of an image showing the extraction results of conference features. In this image, the unique amount limited to conferences with four or less participants is shown by general employee / leader, and the unique amount in all conferences not limited to four or less participants and the unique amount limited to conferences with four or less participants are shown by rank. In this way, the output unit 150 may cause the manager terminal device 200 to display a graph or the like showing the unique amount limited to conferences with four or less participants by general employee / leader.

[0043] Regarding the centrality, the output unit 150 may cause the manager terminal device 200 to display a graph or the like shown in FIG.

[0044] 11 is a diagram showing a third example of an image showing the extraction results of the conference features. In this image, a graph showing the amount of connections is shown by general employee / leader and by rank. The dashed arrows in the figure will be described later. In this way, the output unit 150 may cause the manager terminal device 200 to display a graph showing the amount of connections by general employee / leader and further by rank.

[0045] Returning to FIG. 7, the index value derivation unit 140 organizes the original information for each employee and derives an index value that the employee should aim for (S34). The index value is, for example, based on a comparison between the meeting characteristics of a target employee (for example, an S-rank employee) and the meeting characteristics of a target employee, in order to bring the meeting characteristics of the target employee closer to the meeting characteristics of an S-rank employee. The dashed arrow shown in FIG. 11 indicates the direction of bringing a C-rank employee closer to an S-rank employee, and indicates that, in terms of the degree of connection, it is better to increase the number of meetings in which other employees outside the headquarters participate. In this way, the index value derivation unit 140, for example, calculates the difference between the average value of each meeting characteristic of an S-rank employee and the meeting characteristics of the target employee, and derives the difference as an index value. Since the index value is derived for each meeting characteristic, it is provided to the output unit 150 in the form of a vector having them as elements, for example.

[0046] The output unit 150 outputs the derivation result of the index to the manager terminal device 200, and causes the manager terminal device 200 to display a screen showing the derivation result of the index (S35). FIG. 12 is a diagram showing an example of the screen showing the derivation result of the index. The output unit 150 refers to the index / comment conversion table 176 to acquire a comment to be displayed on the screen. The index / comment conversion table 176 is, for example, information in which the range of indexes for each conference feature is associated with the corresponding comment. The output unit 150 searches the index / comment conversion table 176 for the range of indexes corresponding to the index derived by the index derivation unit 140, and acquires a comment corresponding to the hit range.

[0047] According to the above-described process of [Meeting feature extraction & index value derivation], it is possible to preferably obtain meeting features that indicate the tendency of the target person to participate in the meeting, and further, it is possible to preferably derive the index value that the target person should aim for.

[0048] The above describes the form for carrying out the present invention using an embodiment, but the present invention is not limited to such an embodiment, and various modifications and substitutions can be made within the scope that does not deviate from the gist of the present invention. [Explanation of symbols]

[0049] 10. Terminal device of the person to be managed 50 Calendar Management System 100 Information processing device 110 Acquisition Department 120 Judgment section 130 Meeting feature extraction unit 140 Index value derivation part 150 Output section 200 Administrator terminal device

Claims

1. an acquisition unit that acquires raw information representing the contents of a meeting and including a plurality of pieces of item information; a conference feature extraction unit that extracts conference features from the plurality of item information in the raw information; Equipped with the conference feature extraction unit extracts, as the conference feature, numerical values ​​obtained by converting a plurality of pieces of graph information indicating relationships between the managed persons who participated in the same conference during an observation period; The numerical values ​​of the plurality of pieces of graph information include some or all of the following: a numerical value that increases as the number of other employees connected to an employee increases, a numerical value that increases as the employee is in a position to mediate between different groups, and the number of other employees who are connected to an employee and have a large number of employees connected to the employee. Information processing device.

2. an acquisition unit that acquires raw information representing the contents of a meeting and including a plurality of pieces of item information; a conference feature extraction unit that extracts conference features from the plurality of item information in the raw information; Equipped with the conference feature extraction unit extracts, as the conference feature, a value representing a distribution of affiliation of other managed persons who participated in the same conference as the managed person for each of a plurality of groups in a hierarchical relationship to which the managed person belongs during an observation period; Information processing device.

3. an acquisition unit that acquires raw information representing the contents of a meeting and including a plurality of pieces of item information; a conference feature extraction unit that extracts conference features from the plurality of item information in the raw information; an index value derivation unit that obtains a difference between an average value of the conference features of a target first managed user and the conference features of a second managed user, and derives the difference as an index value to be targeted by the second managed user; An information processing device comprising: Information processing device.

4. An output unit that displays a comment converted from the index value into text on a terminal device.

4. The information processing device according to claim 3.

5. The conference feature extraction unit further extracts a conference volume, a conference number, and / or a conference ratio during an observation period as the conference features.

5. The information processing device according to claim 1.

6. The conference feature extraction unit further extracts, as the conference feature, a unique number of other managed persons participating in the conference together during an observation period.

5. The information processing device according to claim 1.

7. and an output unit that displays a screen in which the extracted conference features are visualized on a terminal device.

3. The information processing device according to claim 1 or 2.

8. An information processing device, A process of obtaining raw information representing the contents of a meeting and including a plurality of items of information; A process of extracting meeting features from the plurality of item information in the raw information; a process of calculating a difference between an average value of the conference feature of a target first managed person and the conference feature of a second managed person, and deriving the difference as an index value to be targeted by the second managed person; An information processing method for performing the above.

9. On the computer, A process of obtaining raw information representing the contents of a meeting and including a plurality of items of information; A process of extracting meeting features from the plurality of item information in the raw information; a process of calculating a difference between an average value of the conference feature of a target first managed person and the conference feature of a second managed person, and deriving the difference as an index value to be targeted by the second managed person; A program for executing the above.

Images