Providing apparatus, providing method, and providing program

Abstract

To effectively evaluate an occurrence risk of a factor that hinders goal achievement.SOLUTION: A provision device 10 acquires each piece of meeting information regarding each of a plurality of meetings held to achieve a specific goal, calculates each analysis value obtained by analyzing each piece of the meeting information of each of the plurality of meetings, evaluates status of the meeting toward achievement of a specific goal based on each analysis value, and notifies administrators or participants of the plurality of meetings of an occurrence of a risk when the risk is detected as the status of the meeting. Further, the provision device 10 calculates the analysis value for each of the plurality of meetings based on a result obtained by inputting each piece of the meeting information into a machine learning model 14b generated by machine learning with the meeting information as an explanatory variable and the analysis value for each attribute of an analysis target as an objective variable.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a providing apparatus, a providing method, and a providing program.

Background Art

[0002] Conventionally, in a meeting, there is a technique for quantitatively evaluating the quality of the meeting. Also, in a meeting, there is a technique that enables quantitative confirmation of the meeting implementation status without bothering the meeting participants or using expensive devices and without worrying about leakage of meeting information.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the above prior art, in a plurality of meetings carried out for achieving a specific goal such as a project, it is difficult to effectively evaluate the occurrence risk of factors that inhibit goal achievement (appropriately, "project risk"). This is because, in the above prior art, although it is possible to evaluate the quality and atmosphere of a single meeting, it is difficult to evaluate the situation, calculate the risk, and notify from a plurality of meetings continuously carried out in a case or project.

[0005] The present invention has been made in view of the above, and an object thereof is to effectively evaluate the occurrence risk of factors that inhibit goal achievement.

Means for Solving the Problems

[0006] The present invention provides a providing apparatus including: an acquisition unit that acquires each conference information regarding each of a plurality of conferences held to achieve a specific goal; a calculation unit that calculates each analysis value obtained by analyzing each conference information of the plurality of conferences; and an evaluation unit that evaluates the status of the conferences toward achieving the specific goal based on each analysis value.

[0007] Further, the present invention provides a providing method in which a computer executes a process of acquiring each conference information regarding each of a plurality of conferences held to achieve a specific goal, calculating each analysis value obtained by analyzing each conference information of the plurality of conferences, and evaluating the status of the conferences toward achieving the specific goal based on each analysis value.

[0008] Further, the present invention provides a providing program for causing a computer to execute a process of acquiring each conference information regarding each of a plurality of conferences held to achieve a specific goal, calculating each analysis value obtained by analyzing each conference information of the plurality of conferences, and evaluating the status of the conferences toward achieving the specific goal based on each analysis value.

Advantages of the Invention

[0009] According to the present invention, there is an effect that the occurrence risk of factors inhibiting goal achievement can be effectively evaluated.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Mode for Carrying Out the Invention

[0011] Hereinafter, a provision device, a provision method, and a provision program according to an embodiment of the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments described below.

[0012] 〔Embodiment〕 Hereinafter, the configuration of the provision system according to the embodiment, the configuration of the provision device, the flow of each process will be described in order, and finally the effects of the embodiment will be described.

[0013] 〔1. Configuration of Provision System 100〕 Using FIG. 1, the configuration of the provision system 100 according to the embodiment will be described in detail. FIG. 1 is a diagram showing a configuration example of the provision system 100 according to the embodiment. Hereinafter, the configuration example of the entire provision system 100, the processing of the provision system 100, and the effects of the provision system 100 will be described in order.

[0014] (1-1. Configuration Example of Entire Provision System 100) The provision system 100 includes a provision device 10, a user terminal 20, a conference information database 30, an analysis information database 40, and a project information database 50. Here, the provision device 10, the user terminal 20, the conference information database 30, the analysis information database 40, and the project information database 50 are communicably connected by wire or wirelessly via a predetermined communication network (network) (not shown).

[0015] The provided system 100 shown in FIG. 1 may include a plurality of providing devices 10, a plurality of user terminals 20, a plurality of conference information databases 30, a plurality of analysis information databases 40, and a plurality of project information databases 50. Also, various databases (conference information database 30, analysis information database 40, project information database 50) may be configured such that two or more databases are integrated. Furthermore, the providing device 10 may be configured to be integrated with the above various databases.

[0016] (1-2. Processing of the Entire Provided System 100) The processing of the entire provided system 100 described above will be explained. Note that the following steps S1 to S7 can also be executed in a different order. Also, some of the following steps S1 to S7 may be omitted.

[0017] (1-2-1. Processing of Step S1) First, the providing device 10 acquires conference information 30a including voice 30a-1, image 30a-2, transcription 30a-3, etc. from the conference information database 30 (step S1). Here, the conference information 30a is information related to a conference and is information that serves as a material for measuring the health state such as the occurrence of risks in a project. For example, the conference information 30a is digital data such as voice, text, video, etc. in a conference, but is not particularly limited. Also, the conference information 30a is information in which information identifying a project such as a project name and a project ID is associated with various digital data.

[0018] (1-2-2. Processing of Step S2) Second, the providing device 10 calculates an analysis value using the acquired meeting information 30a. Here, the analysis value is a numerical value indicating the quality of the meeting calculated by analyzing the meeting information 30a (step S2). For example, as analysis items 14a-1, the providing device 10 analyzes items such as "start to end time", "agenda execution", "sharing of decisions", "status of setting the next meeting", "ratio of speakers", "emotions of speakers", "occurrence status of risks, issues, and negative words", etc., and calculates the analysis value corresponding to the achievement degree of each item as a numerical value from 0 to 100. At this time, the providing device 10 calculates the analysis value using a machine learning model 14b (not shown) generated by machine learning with the meeting information 30a as the explanatory variable and the analysis value for each attribute of the analysis target, which is the analysis item 14a-1, as the objective variable. Also, the providing device 10 may calculate the analysis value based on rules.

[0019] (1-2-3. Processing of step S3) Third, the providing device 10 stores the calculated analysis value in the analysis information database 40 (step S3). At this time, the providing device 10 stores analysis information 40a including the analysis value calculated for each meeting in the analysis information database 40. For example, the providing device 10 stores, as analysis information 40a (40a-1, 40a-2, ···), analysis information 40a-1 such as for a meeting with a "holding date" of "2021 / 10 / 02" as the analysis information of a project with a "project ID" of "P001", where the "time compliance" is "80" and the "agenda execution ratio" is "60", ···, and analysis information 40a-2 such as for a meeting with a "holding date" of "2021 / 11 / 08", where the "time compliance" is "90" and the "agenda execution ratio" is "40", ···.

[0020] (1-2-4. Processing of step S4) Fourth, the providing device 10 acquires the analysis value from the analysis information database 40 (step S4). For example, the providing device 10 acquires analysis information 40a including the analysis value with a "project ID" of "P001" as the designated project.

[0021] (1-2-5. Processing of step S5) Fifthly, the providing device 10 acquires project information 50a from the project information database 50 (step S5). Here, the project information 50a is information including a project ID, an evaluation range, a notification threshold, a notification destination, a notification frequency, etc., which serve as criteria for evaluating a project. For example, the providing device 10 acquires, as the project information 50a, a "project ID" of "P001", an "evaluation range" of "the most recent one month", a "notification destination" of "a@b.com", a "notification threshold" of "70 points or less on average", etc.

[0022] (1-2-6. Processing of step S6) Sixthly, the providing device 10 evaluates the project and detects the occurrence of a project risk (step S6). Here, the providing device 10 evaluates the analysis information 40a according to the criteria of the project information 50a. For example, the providing device 10 calculates the average value of the analysis values of each meeting with a "project ID" of "P001" and uses it as the score value of each meeting. Further, the providing device 10 calculates the average value of the score values of each meeting in the most recent one month, which is the evaluation range, and if it is 70 points or less on average, which is the notification threshold, it detects the occurrence of a risk for the project. At this time, as the notification threshold, in addition to the above average value, the providing device 10 can also set a proportionality constant when linearly approximating the score values of each meeting, and a dead cross between a short-term moving average line and a long-term moving average line based on the score values of each meeting. In addition, the providing device 10 can also acquire the analysis information 40a for each meeting type or period. For example, as the meeting type, the providing device 10 can also acquire the analysis values of meetings belonging to the same project with the same project ID, meetings with high score values to be described later, regular meetings, emergency meetings, etc. In addition, as the meeting period, the providing device 10 can acquire not only the analysis values in a short-term period such as the most recent one month, but also the analysis values in a long-term period such as the most recent one year. A specific example of the risk occurrence detection process by the providing device 10 will be described later.

[0023] (1-2-7. Processing of step S7) Seventh, when the providing device 10 detects the occurrence of a risk, it transmits a risk occurrence notice M to the user terminal 20 (step S7). For example, the providing device 10 notifies the user terminal 20 of the project manager of the project ID "P001" that "a situation with a high likelihood of risk occurrence has been detected in the project ID: P001 in the most recent month." etc. At this time, in addition to the occurrence of a risk, the providing device 10 can also notify advice for improving the project.

[0024] (1-3. Effects of the providing system 100) Hereinafter, as a reference technique, after explaining the problems of techniques for quantitatively evaluating the quality and implementation status of meetings, the effects of the providing system 100 will be explained.

[0025] (1-3-1. Problems) The reference techniques are techniques for quantitatively evaluating the quality of meetings (see, for example, Patent Document 1) and techniques for quantitatively confirming the meeting implementation status without bothering the hands of meeting participants or using expensive devices and without worrying about the leakage of meeting information (see, for example, Patent Document 2), but they have the following problems.

[0026] First, although the above techniques can evaluate the quality and atmosphere of a single meeting, it is difficult to evaluate the situation, calculate risks, and notify them from a plurality of meetings continuously held in a case or project.

[0027] Also, regarding project risks, since people report them through gate reviews, regular reports, etc., information may be modified by human will and there may be a lack of real-time performance.

[0028] (1-3-2. Outline) In the provision system 100, for each of a plurality of meetings held to achieve a specific goal such as a project, meeting information is acquired, analysis values for each of the plurality of meetings obtained by analyzing the respective meeting information are calculated, and based on the calculated analysis values, the status of the meetings toward achieving the specific goal is evaluated. At this time, in the provision system 100, when a risk is detected as the status of a meeting, the occurrence of the risk is notified to the administrators or participants of the plurality of meetings. Also, in the provision system 100, analysis values for each of the plurality of meetings are calculated based on the results obtained by inputting each meeting information into a machine learning model 14b generated by machine learning with the meeting information as an explanatory variable and the analysis value for each attribute to be analyzed as an objective variable.

[0029] That is, in the provision system 100, project risks can be calculated from a plurality of meetings continuously carried out in a project and mechanically conveyed to the project owner or project manager as numerically judged values.

[0030] (1-3-3. Effect) Therefore, the provision system 100 can lead to prompt recovery actions by not waiting for a timing such as a regular report and notifying the administrator of the health status such as the project risk of the project.

[0031] Also, the provision system 100 can eliminate the room for the intervention of human emotions and convey the actual state to the administrator by judging the degree of risk occurrence, which is the health status of the project.

[0032] Also, since the provision system 100 can numerically discriminate risk occurrence factors from a plurality of meetings, it can be utilized as appropriate judgment material regardless of the skills of the recipient.

[0033] Furthermore, the provision system 100 can not only detect project risks but also extract projects with few risk occurrence factors and use them for examples and commendations of good projects.

[0034] As described above, the provision system 100 enables the effective evaluation of the risk of occurrence of factors that inhibit the achievement of a target in a plurality of meetings implemented for the achievement of a specific target.

[0035] [2. Configuration of the provision device 10] Using FIGS. 2 to 5, the functional configurations of the respective devices included in the provision system 100 shown in FIG. 1 will be described. Hereinafter, a configuration example of the provision device 10 according to the embodiment and a specific example of the risk occurrence detection process according to the embodiment will be described in detail.

[0036] (2-1. Configuration example of the provision device 10) First, using FIG. 2, a configuration example of the provision device 10 shown in FIG. 1 will be described. FIG. 2 is a block diagram showing a configuration example of the provision device 10 according to the embodiment. The provision device 10 includes an input unit 11, an output unit 12, a communication unit 13, a storage unit 14, and a control unit 15.

[0037] (2-1-1. Input unit 11) The input unit 11 is in charge of the input of various types of information to the provision device 10. For example, the input unit 11 is realized by a mouse, a keyboard, or the like, and receives the input of setting information and the like to the provision device 10.

[0038] (2-1-2. Output unit 12) The output unit 12 is in charge of the output of various types of information from the provision device 10. For example, the output unit 12 is realized by a display or the like, and outputs the setting information and the like stored in the provision device 10.

[0039] (2-1-3. Communication unit 13) The communication unit 13 is in charge of data communication with other devices. For example, the communication unit 13 performs data communication with each communication device via a router or the like. Further, the communication unit 13 can perform data communication with an operator's terminal (not shown).

[0040] (2-1-4. Storage unit 14) The storage unit 14 stores various information referred to when the control unit 15 operates and various information acquired when the control unit 15 operates. The storage unit 14 has an analysis target attribute storage unit 14a and a machine learning model 14b. Here, the storage unit 14 can be realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, or a storage device such as a hard disk or an optical disk. In the example of FIG. 2, the storage unit 14 is installed inside the providing device 10, but it may be installed outside the providing device 10, or a plurality of storage units may be installed.

[0041] (2-1-4-1. Analysis target attribute storage unit 14a) The analysis target attribute storage unit 14a stores the attributes of the analysis target of the conference information 30a. For example, as analysis items 14a-1 when calculating analysis values for each conference, the analysis target attribute storage unit 14a stores preset analysis items such as "start to end time", "execution of topics", "sharing of decisions", "setting status of the next conference", "ratio of speakers", "emotions of speakers", "occurrence status of risks, issues, and negative words", etc. Further, the analysis target attribute storage unit 14a may store rules regarding the calculation of analysis values for each item of the analysis item 14a-1.

[0042] (2-1-4-2. Machine learning model 14b) The machine learning model 14b is a model trained to output an analysis value in response to the input of the conference information 30a. For example, the machine learning model 14b is a trained model generated by machine learning with the conference information 30a as the explanatory variable and the analysis value for each attribute of the analysis target, which is the analysis item 14a-1, as the objective variable. In the above example, the "explanatory variable" can be digital data that is conference information, and the objective variable can be the number of occurrences or time of the emotion of "anger", which is the emotion of the speaker, and the number of occurrences or appearance rate of negative words such as "impossible". In this case, various parameters of the machine learning model 14b are updated so that the difference between the output value output by the machine learning model 14b in response to the input of the training data and the objective variable becomes small. Also, the machine learning model 14b may be a plurality of trained models generated for each item shown in the analysis item 14a-1.

[0043] (2-1-5. Control Unit 15) The control unit 15 controls the entire providing device 10. The control unit 15 includes an acquisition unit 15a, a calculation unit 15b, an evaluation unit 15c, a notification unit 15d, and an update unit 15e. Here, the control unit 15 can be realized by, for example, an electronic circuit such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit), or an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0044] (2-1-5-1. Acquisition Unit 15a) The acquisition unit 15a acquires each meeting information 30a regarding each of a plurality of meetings held to achieve a specific goal. For example, the acquisition unit 15a acquires digital data including the audio, images, or transcriptions of the plurality of meetings. To explain using a specific example, the acquisition unit 15a acquires, from the meeting information database 30, the meeting information 30a in which the digital data of the audio 30a-1, image 30a-2, and transcription 30a-3 in the meeting with the project ID "P001" and the holding date "2021 / 10 / 02" is associated with the project ID "P001".

[0045] (2-1-5-2. Calculation unit 15b) The calculation unit 15b calculates each analysis value obtained by analyzing each meeting information 30a of each of the plurality of meetings. For example, the calculation unit 15b inputs each meeting information 30a into a machine learning model 14b generated by machine learning with the meeting information 30a as an explanatory variable and the analysis value for each attribute to be analyzed as a target variable, and calculates the analysis value for each of the plurality of meetings based on the obtained result. To explain using a specific example, the calculation unit 15b calculates the time compliance "80", the issue completion rate "60",... which are the analysis values, based on the result obtained by inputting the digital data of the audio 30a-1, image 30a-2, and transcription 30a-3 in the meeting with the project ID "P001" and the holding date "2021 / 10 / 02" into the machine learning model 14b. At this time, the calculation unit 15b calculates the time compliance, which is the analysis value, lower as the number of times the meeting time from the start to the end of the meeting exceeds the scheduled meeting time is larger. On the other hand, the calculation unit 15b calculates the time compliance, which is the analysis value, higher as the number of times the meeting time from the start to the end of the meeting exceeds the scheduled meeting time is smaller.

[0046] Also, regarding the numerical value to be calculated, the calculation unit 15b calculates the analysis value so as to take a numerical value from 0 to 100 according to the quality and achievement degree of the meeting. Further, the calculation unit 13b may calculate the compatibility degree so as to take a numerical value from 0 to 1 according to the quality and achievement degree of the meeting, and the range and unit of the numerical value to be calculated are not particularly limited.

[0047] Further, the calculation unit 15b calculates each analysis value using the digital data of each of the plurality of meetings and stores it in the storage unit. To explain using a specific example, for a plurality of meetings with a project ID of "P001", the calculation unit 15b uses the time compliance rate "80", the progress rate of the agenda "60",... which are the analysis values in the meeting on the holding date "2021 / 10 / 02" as analysis information 40a-1, and also uses the time compliance rate "80", the progress rate of the agenda "60",... which are the analysis values in the meeting on the holding date "2021 / 11 / 08" as analysis information 40a-2, and stores them in the analysis information database 40. At this time, the calculation unit 15b may calculate the average value of the analysis values for each meeting and store it in the analysis information database 40 as analysis information 40a.

[0048] At this time, the calculation unit 15b can calculate the analysis value based on the emotion by waveform analysis of the voice data using the voice 30a-1 in the meeting. Also, the calculation unit 15b can calculate the analysis value based on the emotions such as peace of mind, agitation, anger, etc. by analyzing the image data using the image 30a-2 in the meeting. Further, the calculation unit 15b can calculate the analysis value based on the appearance rate of words associated with project risks by analyzing the character data using the transcription 30a-3 in the meeting. At this time, the calculation unit 15b calculates the progress rate of the agenda, which is the analysis value, to be lower as the number of negative emotions or the number of appearances of words, etc. is larger. On the other hand, the calculation unit 15b calculates the progress rate of the agenda, which is the analysis value, to be higher as the number of negative emotions or the number of appearances of words, etc. is smaller. Furthermore, the calculation unit 15b may extract remarks and emotions that may become harassment from the digital data of the voice 30a-1, the image 30a-2, and the transcription 30a-3 in the meeting and store them in the analysis information database 40 together with the analysis values. As described above, the calculation unit 15b can analyze the atmosphere of the meeting.

[0049] (2-1-5-3. Evaluation Unit 15c) Based on each analysis value, the evaluation unit 15c evaluates the situation of the meeting towards the achievement of a specific goal. For example, the evaluation unit 15c obtains analysis values based on the type or period of the meeting, and uses a predetermined threshold set for each meeting to detect the occurrence of risks as the situation of the meeting. To explain using a specific example, the evaluation unit 15c obtains the analysis values of each meeting with the project ID "P001" from the analysis information database 40, and obtains, as the project information 50a of the project ID "P001" from the project information database 50, an evaluation range of "the most recent one month" and a notification threshold of "average score of 70 points or less", obtains the analysis values of each meeting in the most recent one month of the project ID "P001", calculates the average value of the obtained analysis values as the score value, and when the score value is 70 points or less, detects the occurrence of risks. At this time, as the type of the meeting, the evaluation unit 15c can also obtain and evaluate the analysis values of meetings belonging to the same project with the same project ID, meetings with high score values, regular meetings, emergency meetings, etc. That is, the evaluation unit 15c can obtain and evaluate, from the analysis information database 40, the analysis values calculated from the meeting information 30a by the calculation unit 15b and belonging to the same meeting type or during the evaluation period of the meeting.

[0050] Here, the calculation process of the score value for evaluating the meeting and the project will be described. For example, the evaluation unit 15c calculates the average value of all the analysis values of the specific meeting as the score value for evaluating the meeting. At this time, the evaluation unit 15c may calculate the average value of some of the analysis values of the meeting designated as the evaluation range as the score value, or may calculate the weighted average value as the score value by weighting the items of the analysis item 14a-1 which is the attribute of the analysis target. In addition, the evaluation unit 15c may calculate the average value of all the analysis values of each meeting in order to evaluate the project in which a plurality of meetings have been held, and further calculate the numerical value obtained by averaging the average value by the number of meetings as the score value. At this time, the evaluation unit 15c may calculate the score value from some of the meetings designated as the evaluation range among the plurality of meetings held for the project, or may calculate the score value for each analysis item, or may calculate the weighted average value as the score value by weighting the implemented meetings.

[0051] Regarding the risk occurrence detection process using a threshold value, the evaluation unit 15c calculates the average value of each analysis value in a plurality of meetings including the most recent analysis value as a score value, and when the score value is less than or equal to the threshold value, it detects the occurrence of a risk. Further, the evaluation unit 15c sets the proportionality constant obtained by linear approximation of each analysis value in a plurality of meetings not including the most recent analysis value as the threshold value, calculates the proportionality constant obtained by linear approximation of each analysis value in a plurality of meetings including the most recent analysis value, and when the proportionality constant is less than or equal to the threshold value, it detects the occurrence of a risk. Further, the evaluation unit 15c sets the analysis value at the time when a dead cross occurs between the short-term moving average line and the long-term moving average line of each analysis value in a plurality of meetings as the threshold value, and when the most recent analysis value is less than or equal to the threshold value, it detects the occurrence of a risk. A specific example of the risk occurrence detection process by the evaluation unit 15c will be described later.

[0052] Regarding the extraction process using a score value, the evaluation unit 15c calculates the score values of a plurality of meetings using each analysis value, and extracts a meeting in which the score value is greater than or equal to the threshold value, or a specific target including a predetermined number or more of meetings in which the score value is greater than or equal to the threshold value. Describing using a specific example, the evaluation unit 15c acquires the analysis values of all meetings of the project ID "P002" from the analysis information database 40, calculates the average value of all analysis values of each meeting as the score value, tags the meeting in which the score value is 90 points or more as a "good meeting", and extracts the analysis information 40a of the meeting. Further, the analysis values of all meetings of the project ID "P003" are acquired from the analysis information database 40, the average value of the analysis values of all meetings of the project ID "P003" is calculated as the score value, and when the score value is 90 points or more, the project ID "P003" is tagged as a "good project", and the analysis information 40a of the project is extracted. At this time, the evaluation unit 15c can also display a graph so that the tendency of the score value (appropriately, "score trend") for each evaluation range of the target project ID can be visually grasped.

[0053] Regarding the detection process using keywords and the like, the evaluation unit 15c can detect specific utterances, emotions, etc. using each analysis value. For example, the evaluation unit 15c detects utterances and emotions that may be harassment, which are extracted by the calculation unit 15b, from voice data, image data, and text data in a specific meeting. Also, the evaluation unit 15c detects utterances and emotions that commonly appear in multiple projects from voice data, image data, and text data in meetings belonging to multiple projects. As described above, the evaluation unit 15c can prevent harassment and understand how each project is related.

[0054] (2-1-5-4. Notification Unit 15d) When a risk is detected as the situation of a meeting, the notification unit 15d notifies the occurrence of the risk to the administrators or participants of multiple meetings. Explaining using a specific example, when the score value for the past one month of the project ID "P001" is 70 points or less, which is the notification threshold value, and the occurrence of a project risk is detected, the notification unit 15d sends a risk occurrence notification such as "In the project ID: P001 for the past one month, a situation with a high likelihood of risk occurrence has been detected." to the email address "a@b.com" of the user terminal 20 of the project administrator.

[0055] In addition, when the notification unit 15d detects the occurrence of a risk, it notifies the administrators or participants of multiple meetings of the occurrence of the risk or improvement measures for the multiple meetings. For example, the notification unit 15d notifies advice so that the ongoing project improves as compared with the score trend of the meetings of past projects. Here, "past" refers to past projects conducted during a certain past period, and "ongoing" refers to ongoing projects that have been conducted from a certain past point in time to the present. At this time, the notification unit 15d prepares a template in advance as the content of the advice according to the score trend of the meetings of past projects, and notifies the advice from the prepared template to the ongoing project. For example, the notification unit 15d can use the content of the advice in past projects as the advice for ongoing projects.

[0056] Furthermore, the notification unit 15d notifies the occurrence of a risk to the administrators or participants of multiple meetings based on keywords and the like detected by the evaluation unit 15c. For example, when the frequency or number of occurrences of remarks or emotions that may become harassment detected by the evaluation unit 15c exceeds a predetermined value, the notification unit 15d notifies the occurrence of a risk to the administrators or participants of multiple meetings. As described above, the evaluation unit 15c can execute prevention of harassment and the like in real time.

[0057] (2-1-5-5. Update Unit 15e) When the occurrence of a risk is detected, the update unit 15e updates the type or period for extracting the analysis value or the set threshold value. For example, when the evaluation period of the project in which the occurrence of a risk is detected by the evaluation unit 15c is "the most recent one month", the update unit 15e updates the evaluation period of the project to "the most recent two weeks".

[0058] (2-2. Specific Example of Risk Occurrence Detection Process) Using FIGS. 3 to 5, a specific example of the risk occurrence detection process executed by the evaluation unit 15c of the providing device 10 will be described. FIGS. 3 to 5 are diagrams showing specific examples of the risk occurrence detection process according to the embodiment. Hereinafter, a specific example 1 in which a score value obtained by averaging analysis values is compared with a notification threshold, a specific example 2 in which a proportionality constant when the score values of each meeting are linearly approximated is set as the notification threshold, and a specific example 3 in which a death cross between a short-term moving average line and a long-term moving average line based on the score values of each meeting is set as the notification threshold will be described in this order.

[0059] (2-2-1. Specific Example 1) Using FIG. 3, a specific example 1 in which a score value obtained by averaging analysis values is compared with a notification threshold will be described. In specific example 1, an example will be described in which the most recent one-month meetings in the same project are "Meeting 1" to "Meeting 5", and the notification threshold is set to "70 points".

[0060] The providing device 10 acquires analysis information 40a of the most recent one-month meetings "Meeting 1" to "Meeting 5" in the same project, and calculates the average value of all analysis values of each meeting from "Meeting 1" to "Meeting 5" as the score value of each meeting. In the example of FIG. 3, the providing device 10 calculates that the score value of "Meeting 1" is "60", the score value of "Meeting 2" is "85", the score value of "Meeting 3" is "50", the score value of "Meeting 4" is "70", and the score value of "Meeting 5" is "75". Further, the providing device 10 averages the score values of "Meeting 1" to "Meeting 5" and calculates the average score value as "68". Then, since the calculated average score value "68" of the most recent one-month meetings is less than or equal to the set notification threshold "70", the providing device 10 detects the occurrence of a risk in the project and transmits a risk occurrence notification M to a notification destination such as a project manager.

[0061] In the example of FIG. 3, the average value of the analysis values of each meeting from "Meeting 1" to "Meeting 5" is calculated as the score value of each meeting, but the analysis values of analysis items such as "time compliance" and "issue completion rate" may also be used as the score value of each meeting. Also, a notification threshold may be set for each of the above analysis items.

[0062] (2-2-2. Specific Example 2) Using FIG. 4, a specific example 2 of setting the proportionality constant when linearly approximating the score values of each meeting as the notification threshold will be described. In specific example 2, an example will be described in which the meetings up to one month before excluding the most recent meeting in the same project are "Meeting 1" to "Meeting 5", and the most recent meeting is "Meeting 6".

[0063] The providing device 10 acquires the analysis information 40a of the meetings "Meeting 1" to "Meeting 5" up to one month before excluding the most recent meeting in the same project, and calculates the average value of all the analysis values of each of the meetings "Meeting 1" to "Meeting 5" as the score value of each meeting. In the example of FIG. 4, the providing device 10 calculates that the score value of "Meeting 1" is "50", the score value of "Meeting 2" is "60", the score value of "Meeting 3" is "45", the score value of "Meeting 4" is "65", and the score value of "Meeting 5" is "70". Further, the providing device 10 displays the holding date and time on the horizontal axis and the score values of "Meeting 1" to "Meeting 5" on the vertical axis, linearly approximates them by the least squares method or the like, and sets the slope (proportionality constant) of the approximated straight line as the notification threshold. In the example of FIG. 4, the providing device 10 calculates the proportionality constant of "Meeting 1" to "Meeting 5" as "1.5" and sets it as the notification threshold. Then, the providing device 10 calculates the proportionality constant by linearly approximating the analysis values of a predetermined period of the meetings implemented thereafter in the same manner, and when the proportionality constant is less than or equal to the notification threshold, detects the occurrence of a risk in the project and transmits a risk occurrence notification M to a notification destination such as a project manager. In the example of FIG. 4, when the score value of "Meeting 6", which is the most recent meeting, is "70", the providing device 10 calculates the proportionality constant of "Meeting 1" to "Meeting 6" as "1.1", and since it is less than or equal to the notification threshold of "1.5", transmits a risk occurrence notification M. At this time, the providing device 10 can evaluate that the greater the negative value of the proportionality constant, the greater the risk. In the example of FIG. 4, the providing device 10 may calculate the absolute value "0.4" of the difference in the proportionality constant, which is "-0.4", as a numerical value indicating the magnitude of the risk.

[0064] In the example of FIG. 4, the average value of the analysis values of each of the meetings "Meeting 1" to "Meeting 5" was calculated as the score value of each meeting. However, the analysis value of an analysis item such as "time compliance" or "issue completion rate" may also be used as the score value of each meeting. Further, a proportional constant serving as a notification threshold may be set in advance from the score values of similar projects, or a notification threshold may be set for each of the above analysis items.

[0065] (2-2-3. Specific Example 3) Using FIG. 5, a specific example 3 of setting the death cross between the short-term moving average line and the long-term moving average line based on the score value of each meeting as the notification threshold will be described. In specific example 3, an example of using the short-term moving average line and the long-term moving average line for a predetermined period in the same project will be described.

[0066] The providing device 10 sets the short-term and long-term periods of the moving average line in the same project. Next, the providing device 10 displays the short-term moving average line and the long-term moving average line using the score value of each meeting, and sets the score value at which the death cross has occurred as the notification threshold. Here, the death cross is an intersection point where it is considered that the recent risk has turned downward when the short-term moving average line penetrates the long-term moving average line from above, and indicates the point (intersection point) at which the short-term score value decreases and intersects with the long-term score value. Then, when the death cross occurs, the providing device 10 detects the occurrence of a risk in the project and transmits a risk occurrence notification M to a notification destination such as a project manager.

[0067] Note that the above moving average line may be calculated based on the average value of the analysis values of each meeting using the score value of each meeting, or the analysis value of an analysis item such as "time compliance" or "issue completion rate" may be calculated based on the score value of each meeting.

[0068] [3. Processing Flow of the Providing System 100] Using FIG. 6, the processing flow of the provision system 100 according to the embodiment will be described. FIG. 6 is a flowchart showing an example of the processing flow of the provision process according to the embodiment. Note that the processes of the following steps S101 to S106 can also be executed in a different order. Also, among the processes of the following steps S101 to S106, some processes may be omitted.

[0069] (3-1. Processing of Step S101) First, the provision device 10 identifies the project to be analyzed (step S101). For example, the provision device 10 identifies the project to be analyzed based on the project name or project ID.

[0070] (3-2. Processing of Step S102) Second, the provision device 10 acquires the conference information 30a (step S102). For example, the provision device 10 acquires the digital data of the voice 30a-1, image 30a-2, and transcription 30a-3 in the conference associated with the project ID from the conference information database 30.

[0071] (3-3. Processing of Step S103) Third, the provision device 10 calculates an analysis value from the acquired conference information 30a (step S103). For example, the provision device 10 calculates an analysis value of 0 to 100 for each analysis item, which is an attribute to be analyzed, using the machine learning model 14b.

[0072] (3-4. Processing of Step S104) Fourth, the provision device 10 evaluates the calculated analysis value (step S104). For example, the provision device 10 calculates a score value of the project in the evaluation range using the analysis values of a plurality of conferences conducted for the same project, and evaluates the project.

[0073] (3-5. Processing of Step S105) Fifthly, the providing device 10 compares the score value with a notification threshold (step S105). For example, the providing device 10 averages the score values of the same project within the evaluation range and compares the result with a preset notification threshold. At this time, if the score value is less than or equal to the notification threshold (step S105: Yes), the providing device 10 proceeds to the process of step S106. On the other hand, if the score value is not less than or equal to the notification threshold (step S105: No), the providing device 10 ends the process.

[0074] (3-6. Process of step S106) Sixthly, the providing device 10 notifies the occurrence of a project risk (step S106) and ends the process. For example, when the providing device 10 detects the occurrence of a project risk, it notifies the occurrence of the risk to the user terminals 20 of the project administrators such as the project owner and the project manager.

[0075] [4. Effects of the Embodiment] Finally, the effects of the embodiment will be described. Below, effects 1 to 9 corresponding to the processes according to the embodiment will be described.

[0076] (4-1. Effect 1) First, in the process according to the above-described embodiment, for each of a plurality of meetings held to achieve a specific goal, each meeting information is acquired, each analysis value obtained by analyzing each meeting information of the plurality of meetings is calculated, and based on each analysis value, the situation of the meeting towards achieving the specific goal is evaluated. Therefore, in the process according to the embodiment, in a plurality of meetings implemented to achieve a specific goal, the occurrence risk of factors inhibiting goal achievement can be effectively evaluated.

[0077] (4-2. Effect 2) Second, in the process according to the above-described embodiment, when a risk is detected as the meeting situation, the occurrence of the risk is notified to the administrators or participants of a plurality of meetings. Therefore, in the process according to the embodiment, in a plurality of meetings implemented for achieving a specific goal, the occurrence risk of a factor that inhibits goal achievement can be effectively evaluated, and the evaluation result can be notified to the administrator and participants.

[0078] (4-3. Effect 3) Third, in the process according to the above-described embodiment, for each of a plurality of meetings, an analysis value is calculated based on the result obtained by inputting each meeting information into a machine learning model generated by machine learning with the meeting information as an explanatory variable and the analysis value for each attribute of the analysis target as an objective variable. Therefore, in the process according to the embodiment, in a plurality of meetings implemented for achieving a specific goal, the occurrence risk of a factor that inhibits goal achievement can be effectively and objectively evaluated.

[0079] (4-4. Effect 4) Fourth, in the process according to the above-described embodiment, digital data including the voices, images, or transcriptions of a plurality of meetings is acquired, and each analysis value is calculated using the digital data of each of the plurality of meetings and stored in the storage unit. An analysis value based on the type or period of the meeting is acquired from the storage unit, and the occurrence of a risk is detected as the meeting situation using a predetermined threshold set for each meeting. When the occurrence of a risk is detected, the occurrence of the risk or improvement measures for the plurality of meetings are notified to the administrators or participants of the plurality of meetings. Therefore, in the process according to the embodiment, in a plurality of meetings implemented for achieving a specific goal, the occurrence risk of a factor that inhibits goal achievement can be evaluated more effectively, and the evaluation result can be notified to the administrator and participants more effectively.

[0080] (4-5. Effect 5) Fifthly, in the process according to the above-described embodiment, the average value of each analysis value in a plurality of meetings including the most recent analysis value is calculated as a score value, and when the score value is equal to or less than a threshold value, the occurrence of a risk is detected. Therefore, in the process according to the embodiment, in a plurality of meetings implemented for achieving a specific target, the risk of occurrence of a factor that inhibits target achievement can be effectively evaluated by setting a threshold value using the average value.

[0081] (4-6. Effect 6) Sixthly, in the process according to the above-described embodiment, the proportionality constant obtained by linear approximation of each analysis value in a plurality of meetings is set as a threshold value, and when the proportionality constant including the most recent analysis value is equal to or less than the threshold value, the occurrence of a risk is detected. Therefore, in the process according to the embodiment, in a plurality of meetings implemented for achieving a specific target, the risk of occurrence of a factor that inhibits target achievement can be effectively evaluated by setting a threshold value using the proportionality constant.

[0082] (4-7. Effect 7) Seventhly, in the process according to the above-described embodiment, the analysis value at the time when a death cross occurs based on the short-term moving average line and the long-term moving average line of each analysis value in a plurality of meetings is set as a threshold value, and when the most recent analysis value is equal to or less than the threshold value, the occurrence of a risk is detected. Therefore, in the process according to the embodiment, in a plurality of meetings implemented for achieving a specific target, the risk of occurrence of a factor that inhibits target achievement can be effectively evaluated by setting a threshold value using the death cross.

[0083] (4-8. Effect 8) Eighthly, in the process according to the above-described embodiment, when the occurrence of a risk is detected, the type, period, or set threshold value for extracting the analysis value is updated. Therefore, in the process according to the embodiment, in a plurality of meetings implemented for achieving a specific target, the risk of occurrence of a factor that inhibits target achievement can be effectively evaluated by changing the evaluation range.

[0084] (4-9. Effect 9) Ninthly, in the process according to the above-described embodiment, score values of a plurality of meetings are calculated using each analysis value, and a meeting with a score value equal to or higher than a threshold value, or a specific target including a predetermined number or more of meetings with a score value equal to or higher than the threshold value is extracted. Therefore, in the process according to the embodiment, in a plurality of meetings implemented for achieving a specific target, by extracting objects of high evaluation, it is possible to effectively evaluate the risk of occurrence of factors that inhibit target achievement.

[0085] 〔System〕 Regarding the processing procedures, control procedures, specific names, and information including various data and parameters shown in the above documents and drawings, they can be arbitrarily changed unless otherwise specified.

[0086] In addition, each component of each illustrated device is a functional concept, and it is not necessarily physically configured as shown in the drawing. That is, the specific forms of distribution and integration of each device are not limited to those shown in the drawing. In other words, all or a part of it can be functionally or physically distributed and integrated in arbitrary units according to various loads, usage situations, etc.

[0087] Furthermore, each processing function performed by each device can be realized in whole or in any part by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware by wired logic.

[0088] 〔Hardware〕 Next, a hardware configuration example of the providing device 10 will be described. FIG. 7 is a diagram for explaining a hardware configuration example. As shown in FIG. 7, the providing device 10 includes a communication device 10a, an HDD (Hard Disk Drive) 10b, a memory 10c, and a processor 10d. Also, each part shown in FIG. 7 is mutually connected by a bus or the like.

[0089] The communication device 10a is a network interface card or the like and communicates with other servers. The HDD 10b stores programs and databases for operating the functions shown in FIG. 2.

[0090] The processor 10d reads out a program that executes the same processes as each processing unit shown in FIG. 2 from the HDD 10b or the like and expands it in the memory 10c, thereby operating a process that executes each function described in FIG. 2 and the like. For example, this process executes the same functions as each processing unit included in the providing device 10. Specifically, the processor 10d reads out a program having the same functions as the acquisition unit 15a, the calculation unit 15b, the evaluation unit 15c, the notification unit 15d, the update unit 15e, etc. from the HDD 10b or the like. Then, the processor 10d executes a process that executes the same processes as the acquisition unit 15a, the calculation unit 15b, the evaluation unit 15c, the notification unit 15d, the update unit 15e, etc.

[0091] In this way, the providing device 10 operates as a device that executes various processing methods by reading and executing a program. Further, the providing device 10 can also read out the above program from a recording medium by a medium reading device and execute the read program to realize the same functions as those of the above-described embodiment. Note that the program in this other embodiment is not limited to being executed by the providing device 10. For example, the present invention can be similarly applied when another computer or server executes the program, or when these cooperate to execute the program.

[0092] This program can be distributed via a network such as the Internet. Further, this program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, a MO (Magneto-Optical disk), a DVD (Digital Versatile Disc), etc., and can be executed by being read out from the recording medium by a computer.

Explanation of Reference Numerals

[0093] 10 Providing device 11 Input unit 12 Output unit 13 Communication unit 14 Storage unit 14a Analysis target attribute memory unit 14b Machine learning model 15 Control unit 15a Acquisition unit 15b Calculation unit 15c Evaluation unit 15d Notification unit 15e Update unit 20 User terminal 30 Conference information database 30a Conference information 40 Analysis information database 40a Analysis information 50 Project information database 50a Project information 100 Provision system

Claims

1. An acquisition unit that acquires meeting information for each of a plurality of meetings held to achieve a specific goal; A calculation unit that calculates each analysis value obtained by analyzing the meeting information for each of the plurality of meetings; An evaluation unit that evaluates the status of the meetings towards achieving the specific goal based on the respective analysis values; comprising: The calculation unit: Based on the result obtained by inputting each meeting information into a machine learning model generated by machine learning with the meeting information as an explanatory variable and the analysis value for each attribute to be analyzed as an objective variable, calculates the analysis value for each of the plurality of meetings. A providing device.

2. The evaluation unit: When a risk that inhibits the achievement of the specific goal is detected as the status of the meeting, notifies the occurrence of the risk to the administrators or participants of the plurality of meetings. The providing device according to claim 1.

3. The acquisition unit: Acquires digital data including the audio, images, or transcriptions of the plurality of meetings; The calculation unit: Uses the digital data for each of the plurality of meetings to calculate each analysis value and stores it in a storage unit; The evaluation unit: Acquires an analysis value based on the type or period of the meeting from the storage unit, and uses a predetermined threshold set for each meeting to detect the occurrence of a risk that inhibits the achievement of the specific goal; When detecting the occurrence of the risk, notifies the occurrence of the risk or improvement measures for the plurality of meetings to the administrators or participants of the plurality of meetings. The providing device according to claim 1 or 2.

4. The evaluation unit: Calculates the average value of each analysis value in the plurality of meetings including the most recent analysis value as a score value, and when the score value is less than or equal to the threshold, detects the occurrence of the risk. The providing device according to claim 3.

5. The evaluation unit: Sets the proportionality constant obtained by linear approximation of each analysis value in the plurality of meetings not including the most recent analysis value to the threshold, calculates the proportionality constant including the most recent analysis value, and when the proportionality constant including the most recent analysis value is less than or equal to the threshold, detects the occurrence of the risk. The providing device according to claim 3.

6. The evaluation unit: Sets the analysis value at the time when a death cross occurs between the short-term moving average line and the long-term moving average line of each analysis value in the plurality of meetings as the threshold, and when the most recent analysis value is less than or equal to the threshold, detects the occurrence of the risk. The providing device according to claim 4.

7. When the occurrence of the risk is detected, an update unit that updates the type or period for extracting the analysis value or the set threshold value. The providing apparatus according to any one of claims 3 to 6, further comprising the update unit.

8. The evaluation unit calculates a score value for each of the plurality of meetings using the respective analysis values, and extracts a meeting in which the score value is equal to or greater than a threshold value, or the specific target in which the meeting in which the score value is equal to or greater than the threshold value includes a predetermined number or more. The providing apparatus according to claim 1.

9. A computer acquires each meeting information regarding each of a plurality of meetings held to achieve a specific target, calculates each analysis value obtained by analyzing each meeting information of each of the plurality of meetings, evaluates the situation of the meetings towards achieving the specific target based on each analysis value, executes a process, and calculates the analysis values of each of the plurality of meetings based on the results obtained by inputting each meeting information into a machine learning model generated by machine learning using the meeting information as an explanatory variable and the analysis values for each attribute to be analyzed as an objective variable. A providing method.

10. A computer is caused to acquire each meeting information regarding each of a plurality of meetings held to achieve a specific target, calculate each analysis value obtained by analyzing each meeting information of each of the plurality of meetings, evaluate the situation of the meetings towards achieving the specific target based on each analysis value, execute a process, and calculate the analysis values of each of the plurality of meetings based on the results obtained by inputting each meeting information into a machine learning model generated by machine learning using the meeting information as an explanatory variable and the analysis values for each attribute to be analyzed as an objective variable. A providing program.

Images