Information processing device, information processing system, information processing method, and program

Abstract

An information processing device, an information processing system, an information processing method, and a program are provided that can improve the level of understanding of users participating in a Web conference. [Solution] In an information processing device, a data analysis means analyzes reaction data indicating reactions of users participating in a Web conference. A level of understanding calculation means calculates the level of understanding of the users based on the analysis result of the reaction data. A question generation means generates a question for confirming understanding of a word. A question sending means sends the question to a user terminal of a user whose level of understanding is lower than a preset standard. An answer receiving means receives an answer to the question from the user terminal. An answer processing means generates information about the word based on the answer, and sends the information to the user terminal of the user whose level of understanding is lower than the standard.

Description

[Technical field]

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

[0002] In recent years, the number of web conferences has increased with the spread of telework. Compared to face-to-face meetings, online meetings lack information such as the atmosphere of the room, the gaze of participants, and the attitude of participants. This makes it easy for participants to not get a chance to speak. When words whose meanings are unclear (hereafter referred to as unknown words) appear, participants are unable to understand the content of the meeting, which causes a problem of the quality of the meeting declining.

[0003] If the unknown word is a common term, the participant can research it by themselves. However, during the research, the participant cannot join the meeting, and the meeting proceeds without understanding the content. Also, if the unknown word is an internal company term, the participant can ask other participants. However, the problem is that participants' questions can disrupt the progress of the meeting.

[0004] For example, Patent Document 1 describes a system that assists a user who is viewing educational content. The system estimates the user's level of understanding of the content based on the user's eye movements and outputs assist information to promote the user's understanding. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent Publication No. 2022-58315 Summary of the Invention [Problem to be solved by the invention]

[0006] The system described in Patent Document 1 assists users who view educational content. However, this system does not assist users who are participating in a Web conference. In a Web conference, participants speak in real time, but the speech of the participants differs from the content prepared in advance.

[0007] An object of the present disclosure is to provide an information processing device, an information processing system, an information processing method, and a program that solve the above-mentioned problems. [Means for solving the problem]

[0008] The information processing device according to one embodiment of the present disclosure includes a data analysis means for receiving reaction data indicating reactions of users participating in a Web conference from user terminals used by the users in the Web conference, and analyzing the reaction data. a data analysis means for receiving video data generated by a camera included in the user terminal as the reaction data and analyzing the user's emotions based on the video data, or for receiving audio data generated by a microphone included in the user terminal as the reaction data and analyzing a ratio of the user's speaking time to the elapsed conference time based on the audio data, or for receiving input data inputted into the user terminal by the user as the reaction data and analyzing a ratio of the user's non-input time to the elapsed conference time based on the input data. and, a value obtained by combining at least two of the value indicating the emotion, the value indicating the ratio of the speech time, and the value indicating the ratio of the non-input time; The user's level of understanding As a question generating means for generating a question for confirming understanding of words used in the Web conference; a question sending means for sending the question to a user terminal of a user whose understanding level is lower than a preset standard; an answer receiving means for receiving an answer to the question from the user terminal; If it is wrong The said word Show the meaning of and a response processing means for generating information and transmitting the information to the user terminal of the user whose level of understanding is lower than the standard.

[0009] An information processing system according to one aspect of the present disclosure includes the above-mentioned information processing device, and the user terminal that transmits the reaction data and the answer to the information processing device and receives the question and the information from the information processing device.

[0010] An information processing method according to an embodiment of the present disclosure includes receiving reaction data indicating reactions of users participating in a Web conference from user terminals used by the users in the Web conference; receiving video data generated by a camera included in the user terminal as the reaction data, and analyzing the emotion of the user based on the video data; or receiving audio data generated by a microphone included in the user terminal as the reaction data, and analyzing a ratio of speech time to elapsed conference time of the user based on the audio data; or receiving input data inputted into the user terminal by the user as the reaction data, and analyzing a ratio of non-input time to elapsed conference time of the user based on the input data; and calculating a value indicating the emotion, a value indicating the ratio of speech time, and a value indicating the ratio of non-input time, or a combined value of at least two of these. The user's level of understanding Asa user terminal of a user having a lower level of understanding than a preset standard; a user terminal of a user having a lower level of understanding than a preset standard; If it is wrong The said word Show the meaning of Information is generated and transmitted to the user terminal of the user whose level of understanding is lower than the standard.

[0011] A program according to one embodiment of the present disclosure includes: The information processing method Execute the command. Effect of the Invention

[0012] According to the above aspect, it is possible to improve the level of understanding of users participating in a Web conference. [Brief description of the drawings]

[0013] [Figure 1] 1 is a block diagram of a Web conference system according to the present disclosure. [Diagram 2] FIG. 2 is a diagram illustrating an example of employee information stored in an employee information database (DB) according to the present disclosure. [Diagram 3] 1 is a diagram showing an example of understanding level information held in an understanding level DB according to the present disclosure; [Figure 4] 1 is a diagram showing an example of word information held in a word DB according to the present disclosure. [Diagram 5] 2 is a hardware configuration diagram of a Web conference server according to the present disclosure. [Figure 6] FIG. 2 is a hardware configuration diagram of a DB server according to the present disclosure. [Figure 7] FIG. 2 is a hardware configuration diagram of a user terminal according to the present disclosure. [Figure 8] FIG. 2 is a sequence diagram of a Web conference system according to the present disclosure. [Figure 9] FIG. 2 is a sequence diagram of a Web conference system according to the present disclosure. [Figure 10] FIG. 2 is a sequence diagram of a Web conference system according to the present disclosure. [Figure 11]5 is a flowchart showing the operation of a Web conference server according to the present disclosure. [Figure 12] 4 is a schematic diagram showing the operation of a Web conference server according to the present disclosure. FIG. [Figure 13] FIG. 13 illustrates an example of understanding according to the present disclosure. [Figure 14] FIG. 2 is a diagram illustrating an example of audio data according to the present disclosure. [Figure 15] FIG. 2 illustrates an example of a keyboard input according to the present disclosure. [Figure 16] 1 is a diagram illustrating an example of employee information stored in an employee information DB according to the present disclosure. [Figure 17] FIG. 13 is a diagram illustrating an example of a correction value for each position according to the present disclosure. [Figure 18] 1 is a diagram showing an example of word information held in a word DB according to the present disclosure. [Figure 19] FIG. 2 is a diagram showing an example of a display screen of a user terminal according to the present disclosure. [Figure 20] FIG. 2 is a diagram showing an example of a display screen of a user terminal according to the present disclosure. [Figure 21] FIG. 13 illustrates an example of understanding according to the present disclosure. [Figure 22] 1 is a diagram showing an example of understanding level information held in an understanding level DB according to the present disclosure; [Figure 23] FIG. 13 illustrates an example of understanding according to the present disclosure. [Figure 24] FIG. 13 illustrates an example of understanding according to the present disclosure. [Diagram 25] 1 is a diagram showing an example of understanding level information held in an understanding level DB according to the present disclosure; [Figure 26] FIG. 13 illustrates an example of understanding according to the present disclosure. [Figure 27] FIG. 13 illustrates an example of understanding according to the present disclosure. [Figure 28] 1 is a diagram showing an example of understanding level information held in an understanding level DB according to the present disclosure; [Figure 29] 1 is a block diagram of a Web conference system according to the present disclosure. [Diagram 30]5 is a flowchart showing the operation of a Web conference server according to the present disclosure. [Diagram 31] FIG. 2 is a diagram showing an example of a display screen of a user terminal according to the present disclosure. [Diagram 32] FIG. 2 is a diagram showing an example of a display screen of a user terminal according to the present disclosure. [Diagram 33] FIG. 2 is a diagram showing an example of a display screen of a user terminal according to the present disclosure. [Diagram 34] FIG. 2 is a diagram showing an example of a display screen of a user terminal according to the present disclosure. [Diagram 35] FIG. 2 is a diagram showing an example of a display screen of a user terminal according to the present disclosure. [Diagram 36] 1 is a diagram showing an example of word information held in a word DB according to the present disclosure. [Figure 37] FIG. 1 is a block diagram of an information processing device according to the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, each embodiment will be described with reference to the drawings. In all the drawings, the same or corresponding components are denoted by the same reference numerals, and common descriptions will be omitted.

[0015] <First embodiment> An embodiment of the present disclosure will be described below with reference to the drawings. FIG. 1 shows a functional configuration of a Web conference system 100 of this embodiment. The Web conference system 100 calculates each user's level of understanding of the conference contents in real time, and asks questions to users with low levels of understanding to check their understanding of words used in the conference. The Web conference system 100 provides users with information that promotes understanding based on the answers to the questions.

[0016] The Web conference system 100 can automatically ask questions to users and obtain answers. Therefore, the Web conference system 100 can improve the user's understanding and improve the quality of the conference without disrupting the progress of the conference by having the user investigate the issue himself or ask questions to others.

[0017] The Web conference system 100 includes a Web conference server 1, a database (DB) server 2, and a user terminal 3. The Web conference server 1 and the DB server 2 are computer systems such as a personal computer (PC). The user terminal 3 is an information device such as a PC, a smartphone, or a tablet terminal.

[0018] The Web conference server 1 is connected to a network and relays the Web conference. The DB server 2 holds the contents processed by the Web conference server 1. The user terminal 3 communicates with the Web conference server 1 via the network. The Web conference system 100 includes multiple user terminals 3, but the configuration of one user terminal 3 is shown in FIG.

[0019] The Web conference server 1 includes a user identification unit 11, a sentiment analysis unit 12, a speech volume analysis unit 13, a keyboard input analysis unit 14, an answer processing unit 15, an understanding level calculation unit 16, a question generation unit 17, and a question transmission unit 18. The DB server 2 includes an employee information DB 21, an understanding level DB 22, and a vocabulary DB 23. The user terminal 3 includes a camera 31, a microphone 32, a keyboard 33, and a question and answer unit 34.

[0020] The user identification unit 11 has the following functions: The user identification unit 11 refers to the employee information DB 21 of the DB server 2, and identifies users (participants of the Web conference) stored in the employee information DB 21. The user identification unit 11 acquires employee information of users participating in the Web conference from the employee information DB 21.

[0021] The emotion analysis unit 12 has the following functions. The emotion analysis unit 12 receives video data generated by the camera 31 of the user terminal 3 from the user terminal 3. The emotion analysis unit 12 analyzes the degree of each emotion of [joy, trust, fear, surprise, sadness, disgust, anger, and expectation] based on the video data. The emotion analysis unit 12 transmits the analysis result to the understanding degree calculation unit 16.

[0022] The speech amount analysis unit 13 has the following functions. The speech amount analysis unit 13 receives voice data generated by the microphone 32 of the user terminal 3 from the user terminal 3. The speech amount analysis unit 13 analyzes the amount of speech of the user by analyzing the length of the speech section. The speech amount analysis unit 13 transmits the analysis result to the understanding level calculation unit 16. The speech amount analysis unit 13 extracts frequently occurring words from the voice data received from all user terminals 3, and transmits the frequently occurring words to the question generation unit 17.

[0023] The keyboard input analysis unit 14 has the following functions. The keyboard input analysis unit 14 receives input data generated by the keyboard 33 of the user terminal 3 from the user terminal 3. The keyboard input analysis unit 14 analyzes the length of keystroke time to analyze the input frequency of the user. The keyboard input analysis unit 14 transmits the analysis result to the understanding degree calculation unit 16.

[0024] The understanding degree calculation unit 16 has the following functions. The understanding degree calculation unit 16 receives data sent by the emotion analysis unit 12, the comment volume analysis unit 13, the keyboard input analysis unit 14, and the answer processing unit 15, and calculates the understanding degree of the users participating in the Web conference. The understanding degree calculation unit 16 registers the calculated understanding degree in the understanding degree DB 22 of the DB server 2. The understanding degree calculation unit 16 determines a threshold value from the understanding degrees of all participants, and selects a target user to whom a question is to be sent. The understanding degree calculation unit 16 transmits information on the user terminal 3 of the target user to the question transmission unit 18. The understanding degree calculation unit 16 updates the user's understanding degree based on the correct / incorrect data sent by the answer processing unit 15. The correct / incorrect data indicates whether the user's answer is correct or not. The understanding degree calculation unit 16 registers the updated understanding degree in the understanding degree DB 22 of the DB server 2.

[0025] The question generation unit 17 has the following functions. The question generation unit 17 receives frequently occurring words from the comment volume analysis unit 13. The question generation unit 17 refers to the word DB 23 of the DB server 2, and acquires the meanings of the frequently occurring words received from the comment volume analysis unit 13 from the word DB 23. The question generation unit 17 generates question data which is a character string of a question sentence including the frequently occurring words received from the comment volume analysis unit 13. The question generation unit 17 transmits the generated question data to the question sending unit 18. The question generation unit 17 generates correct answer data paired with the question data by using the meanings of the frequently occurring words, and transmits the correct answer data to the answer processing unit 15.

[0026] The question sending unit 18 has the following functions: The question sending unit 18 receives information about the user terminal 3 of the target user who is to send a question from the understanding level calculation unit 16, and receives question data from the question generation unit 17. The question sending unit 18 sends the question data to the question answering unit 34 of the user terminal 3 of the target user.

[0027] The answer processing unit 15 has the following functions. The answer processing unit 15 receives answer data sent by the question answering unit 34 of the user terminal 3, and receives correct answer data sent by the question generation unit 17. The answer processing unit 15 generates correct / incorrect data based on the answer data and correct answer data, and transmits the correct / incorrect data to the understanding level calculation unit 16 and the question answering unit 34. The answer processing unit 15 transmits the meanings of frequently occurring words used to generate the question data to the question answering unit 34.

[0028] The employee information DB 21 holds employee information including the employee's job title. Fig. 2 shows an example of employee information held in the employee information DB 21. The employee information includes each user's ID 211, employee name 212, and job title 213. The employee information DB 21 may be provided by an external service.

[0029] The understanding level DB 22 holds information on the level of understanding of each user regarding a conference. Fig. 3 shows an example of the understanding level information held in the understanding level DB 22. The understanding level information includes an ID 221, an employee name 222, and an understanding level 223 of each user.

[0030] The word DB 23 holds word information. Fig. 4 shows an example of word information held in the word DB 23. The word information includes an ID 231, a word 232, and a meaning 233.

[0031] The camera 31, microphone 32, and keyboard 33 generate reaction data indicative of the reactions of users participating in the Web conference.

[0032] The camera 31 has the following functions: The camera 31 captures the faces of users participating in a Web conference using the user terminal 3, and generates video data as reaction data. The user terminal 3 transmits the video data to the emotion analysis unit 12 of the Web conference server 1.

[0033] The microphone 32 has the following functions. The microphone 32 collects the voices of users participating in a Web conference using the user terminal 3, and generates voice data as reaction data. The user terminal 3 transmits the voice data to the speech volume analysis unit 13 of the Web conference server 1.

[0034] The keyboard 33 has the following functions. The keyboard 33 is an input device used by users participating in a Web conference to input information. The keyboard 33 generates input data as reaction data based on information input by the user. For example, the input data indicates a character string input by the user. The user terminal 3 transmits the input data to the keyboard input analysis unit 14 of the Web conference server 1.

[0035] The user terminal 3 may include an input device other than the keyboard 33. For example, the user terminal 3 may include a button, a touch screen, or the like as an input device.

[0036] The question answering unit 34 has the following functions. The question answering unit 34 receives question data sent by the question sending unit 18 of the Web conference server 1, and displays the question and two-choice answer buttons on the display 309. When the user presses the two-choice answer button, the question answering unit 34 sends answer data indicating the user's answer to the answer processing unit 15 of the Web conference server 1. The question answering unit 34 receives the correct / incorrect data and the meanings of frequently occurring words sent by the answer processing unit 15. The question answering unit 34 displays whether the user's answer is correct or not, and the meanings of the frequently occurring words on the display 309.

[0037] 5 shows the hardware configuration of the Web conference server 1. The Web conference server 1 is a computer equipped with various hardware components such as a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, other storage devices 104, and a communication device 105. The CPU 101 executes a program stored in the ROM 102 to realize the functions of a user identification unit 11, a sentiment analysis unit 12, a comment volume analysis unit 13, a keyboard input analysis unit 14, a response processing unit 15, an understanding level calculation unit 16, a question generation unit 17, and a question transmission unit 18. The communication device 105 communicates with the DB server 2 and the user terminal 3.

[0038] 6 shows the hardware configuration of the DB server 2. The DB server 2 is a computer equipped with various hardware components such as a CPU 201, a ROM 202, a RAM 203, another storage device 204, and a communication device 205. The storage device 204 holds an employee information DB 21, an understanding level DB 22, and a vocabulary DB 23. The communication device 105 communicates with the Web conference server 1.

[0039] 7 shows the hardware configuration of the user terminal 3. The user terminal 3 is a computer equipped with various hardware components such as a CPU 301, a ROM 302, a RAM 303, another storage device 304, a communication device 305, a camera 306, a microphone 307, a keyboard 308, and a display 309. The CPU 301 executes a program stored in the ROM 302 to realize the function of a question and answer unit 34. The communication device 305 executes communication with the Web conference server 1. The camera 306 corresponds to the camera 31. The microphone 307 corresponds to the microphone 32. The keyboard 308 corresponds to the keyboard 33. The display 309 displays question data and the like.

[0040] The programs for implementing the functions of the Web conference server 1, DB server 2, and user terminal 3 are recorded on a computer-readable recording medium (ROM in the above example). A "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into computer systems. Furthermore, "computer-readable recording medium" also includes devices that hold programs for a certain period of time, such as volatile memory (RAM) inside a computer system that becomes a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line.

[0041] The above program may also be transmitted from a computer system in which the program is stored in a storage device or the like to another computer system via a transmission medium, or by transmission waves in the transmission medium. Here, the "transmission medium" that transmits the program refers to a medium that has the function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The above program may also be for realizing part of the above-mentioned functions. Furthermore, it may be a so-called difference file (difference program) that can realize the above-mentioned functions in combination with a program already recorded in the computer system.

[0042] A description will now be given of the operation of the Web conference system 100. Figures 8, 9, and 10 show an example of the operation of the Web conference system 100. The order in which the processes shown in Figures 8, 9, and 10 are executed is merely an example, and the order is not limited thereto.

[0043] The camera 31 of the user terminal 3 generates video data, and the user terminal 3 transmits the video data generated by the camera 31 to the emotion analysis unit 12 of the Web conference server 1 (step S301). The emotion analysis unit 12 receives the video data (step S101). The transmission and reception of the video data are repeated.

[0044] The microphone 32 of the user terminal 3 generates voice data, and the user terminal 3 transmits the voice data generated by the microphone 32 to the speech amount analysis unit 13 of the Web conference server 1 (step S302). The speech amount analysis unit 13 receives the voice data (step S102). The transmission and reception of the voice data are repeated.

[0045] The keyboard 33 of the user terminal 3 generates input data, and the user terminal 3 transmits the input data generated by the keyboard 33 to the keyboard input analysis unit 14 of the Web conference server 1 (step S303). The keyboard input analysis unit 14 receives the input data (step S103). The transmission and reception of the input data are repeated.

[0046] The emotion analysis unit 12 processes the received video data and analyzes the user's emotions. Usually, when the user understands the contents of the conference, positive emotions among joy, anger, sadness, and happiness are strongly expressed. Conversely, when the user does not understand the contents of the conference, negative emotions are strongly expressed. The emotion analysis unit 12 transmits the analysis result to the understanding degree calculation unit 16 (step S104).

[0047] The speech amount analysis unit 13 processes the received voice data and analyzes the amount of speech made by the user. Specifically, the speech amount analysis unit 13 calculates the time that the microphone 32 is unmuted relative to the elapsed time of the conference, and calculates the user's participation rate in the conference. If the user actively speaks in the conference, the speech amount analysis unit 13 determines that the user has a sufficient grasp of the contents of the conference and has a high level of understanding. The speech amount analysis unit 13 transmits the analysis result to the understanding level calculation unit 16 (step S105).

[0048] The keyboard input analysis unit 14 processes the received input data and analyzes the frequency of input by the user. If there is little keyboard input during a meeting, the user is concentrating on the meeting. In such a case, the keyboard input analysis unit 14 determines that the level of understanding is high. If there is a lot of keyboard input during a meeting, the user may be performing a task other than the content of the meeting and may not be concentrating on the meeting. In such a case, the keyboard input analysis unit 14 determines that the level of understanding is low. The keyboard input analysis unit 14 transmits the analysis result to the understanding calculation unit 16 (step S106).

[0049] Every five minutes after the start of the conference, the emotion analysis unit 12, the speech volume analysis unit 13, and the keyboard input analysis unit 14 execute the above-mentioned processes. The emotion analysis unit 12 generates an emotion analysis value E t The speech volume analysis unit 13 calculates the speech detection value V t The keyboard input analysis unit 14 calculates the non-keystroke detection value K t Calculate.

[0050] The order of steps S104, S105, and S106 is not limited to the order shown in Fig. 8. Steps S104, S105, and S106 can be executed in any order.

[0051] In step S104, the emotion analysis unit 12 calculates the emotion analysis value E tThe details of the process of calculating emotion will be described. Although various emotions are output by the learning model, an example using eight emotions based on Plutchik's wheel of emotions will be described in this embodiment. However, the same calculation method as this embodiment can be applied to a learning model that outputs emotions other than the eight emotions.

[0052] Plutchik's Emotion Wheel is an emotion model that shows that many emotions can be expressed by combining eight basic emotions: "happiness," "trust," "fear," "surprise," "sadness," "disgust," "anger," and "anticipation." Of these basic emotions, "happiness" and "trust" are classified as positive emotions, "sadness," "disgust," "anger," and "fear" are classified as negative emotions, and "surprise" and "anticipation" are classified as neutral emotions.

[0053] When a user expresses positive emotions, the level of understanding is defined as high, when a user expresses negative emotions, the level of understanding is low, and when a user expresses neutral emotions, the level of understanding is medium. In addition, the sentiment analysis value is in the range of 0 to 100, and the higher the number, the higher the level of understanding.

[0054] The degree of understanding of a positive emotion is defined as 100, the degree of understanding of a neutral emotion as 50, and the degree of understanding of a negative emotion as 0. The emotion analysis unit 12 calculates the sum of the probabilities classified into each emotion, multiplies the degree of understanding corresponding to each emotion by the sum of each probability, and calculates the sum of the results. t is expressed by the following equation (1). E t = 100 × (sum of positive emotions) + 50 × (sum of neutral emotions) + 0 × (sum of negative emotions) (1)

[0055] For example, if the results of the sentiment analysis are [Happiness 80%, Trust 10%, Fear 1%, Surprise 5%, Sadness 0.5%, Disgust 0.5%, Anger 0.5%, Anticipation 2.5%], then classifying the results of the sentiment analysis into positive, negative, and neutral would look like this: Positive emotion: Happiness (80%) + Trust (10%) = 90 Neutral emotion: Surprise (5%) + Anticipation (2.5%) = 7.5 Negative emotions: Sadness (0.5%) + Disgust (0.5%) + Anger (0.5%) + Fear (1%) = 2.5 Sentiment analysis value: 100 x 90 / 100 + 50 x 7.5 / 100 + 0 x 2.5 / 100 = 93.75

[0056] By rounding off the decimal point of the sentiment analysis value, the sentiment analysis unit 12 obtains the sentiment analysis value E t We get 94 as

[0057] In step S105, the speech volume analysis unit 13 calculates the speech detection value V t The details of the process of calculating the voice detection value V are described below. The time when the microphone 32 is unmuted is defined as the same as the user's speaking time. Also, in a conference, it is defined that a person who speaks a lot understands the contents of the conference, and a person who speaks little does not understand the contents of the conference. The ratio of the cumulative time that the user spoke to the elapsed conference time is defined as the voice detection value V. t The voice detection value V t is expressed by the following equation (2).

[0058]

number

[0059] For example, if the speaking time is 10 minutes in a 30-minute meeting, the voice detection value V t becomes 33.

[0060] In step S106, the keyboard input analysis unit 14 detects the non-keystroke detection value K t The details of the process of calculating are described below. When there is a lot of keyboard input during a conference, the user may be performing a task other than the content of the conference. Therefore, when there is a lot of keyboard input during a conference, it is defined that the user has a low level of understanding of the conference. On the other hand, when there is little keyboard input, it is defined that the user has a high level of understanding of the conference.

[0061] The keyboard input analysis unit 14 monitors the state of the keyboard 33 based on the input data. If there is no keyboard input for one second or more, the keyboard input analysis unit 14 determines that the state of the keyboard 33 is a "no keyboard input" state and starts measuring time. When keyboard input is next detected, the keyboard input analysis unit 14 ends measuring time. The keyboard input analysis unit 14 calculates the ratio of the cumulative time of this "no keyboard input" state to the elapsed time of the conference as a non-keystroke detection value K. t The non-touch detection value K t is expressed by the following equation (3).

[0062]

number

[0063] For example, in a 30-minute meeting, if the time spent typing on the keyboard by the user is 20 minutes, the time spent not typing on the keyboard by the user is 10 minutes. In this case, the non-keystroke detection value K is calculated from the value of formula (3) (10 / 30=33%). t becomes 33.

[0064] The understanding degree calculation unit 16 calculates the emotion analysis value E t is received from the emotion analysis unit 12, and the voice detection value V t is received from the speech volume analysis unit 13, and the non-keystroke detection value K t from the keyboard input analysis unit 14. The understanding degree calculation unit 16 receives the emotion analysis value E t , voice detection value V t , and the non-touch detection value K t is expressed as a measurement vector, and the absolute value of the measurement vector is the measurement M t (Step S107). The measured value M t is expressed by the following equation (4).

[0065]

number

[0066] It is assumed that the higher the position of a conference participant, the greater the amount of prior knowledge. Therefore, the understanding level calculation unit 16 calculates the measurement value M t In this embodiment, the understanding level calculation unit 16 uses the job title of the employee as the attribute of the employee.

[0067] The user identification unit 11 accesses the employee information DB 21 of the DB server 2. The user identification unit 11 holds the ID of each user participating in the Web conference, and transmits the ID to the DB server 2. The DB server 2 acquires the job title of the user having the ID received from the user identification unit 11 from the employee information DB 21, and transmits the job title to the user identification unit 11 (step S211). The user identification unit 11 receives the job title transmitted by the DB server 2, and transmits the job title and the above ID to the understanding level calculation unit 16 (step S111).

[0068] The understanding level calculation unit 16 accesses the understanding level DB 22 of the DB server 2. At this time, the understanding level calculation unit 16 transmits the ID received from the user identification unit 11 to the DB server 2. The DB server 2 acquires the understanding level of the user having the ID received from the understanding level calculation unit 16 from the understanding level DB 22, and transmits the understanding level to the user identification unit 11 (step S212). The understanding level calculation unit 16 receives the understanding level transmitted by the DB server 2 (step S112).

[0069] The understanding level DB 22 holds the understanding level previously calculated by the understanding level calculation unit 16. Before the understanding level calculation unit 16 calculates the understanding level for the first time, the understanding level held in the understanding level DB 22 is an initial value (for example, 0).

[0070] The understanding degree calculation unit 16 calculates the measurement value M t The understanding level calculation unit 16 adds a position correction value C according to the position to the measurement value M t By adding the previous understanding level to each user's understanding level J, t is calculated (step S113).

[0071] Comprehension level J for evaluationt is the previous understanding level P t-1 , the measured value M t By using the position correction value C (first time only), it is expressed by the following equation (5). J t =P t-1 +M t +C(First Time Only) ···(5)

[0072] The understanding degree calculation unit 16 calculates the understanding degree for judgment J t Specifically, the understanding degree calculation unit 16 multiplies the median of the understanding degrees for judgment of all users participating in the conference by a magnification factor to obtain the threshold value Th t For example, the magnification is 0.7, but is not limited to this. t Determine the level of understanding J t is the threshold Th t The understanding level calculation unit 16 selects a user who has a lower understanding level than the target user as a target user. The understanding level calculation unit 16 transmits information for identifying the user terminal 3 of the target user to the question transmission unit 18 (step S114).

[0073] The speech amount analysis unit 13 performs voice recognition on the voice data received from the user terminal 3 and converts the voice data into character string data. Thereafter, the speech amount analysis unit 13 counts the number of times each word appears at five-minute intervals and extracts the word that appears most frequently in the five-minute period (frequent word). The speech amount analysis unit 13 executes this process by using the voice data received from the user terminals 3 of all users participating in the conference. The frequent word is the word that appears most frequently in the speech of all users participating in the conference. The speech amount analysis unit 13 transmits the extracted frequent word to the question generation unit 17 (step S115).

[0074] The question generator 17 generates question data and answer data (step S116) by executing the process shown in Fig. 11. Details of step S116 will be described with reference to Fig. 11.

[0075] The question generation unit 17 accesses the word DB 23 of the DB server 2. At this time, the question generation unit 17 transmits the frequently occurring words received from the comment amount analysis unit 13 to the DB server 2 (step S1161). The DB server 2 obtains the meanings of the frequently occurring words received from the question generation unit 17 from the word DB 23, and transmits the meanings to the question generation unit 17 (step S213 in FIG. 9).

[0076] The question generation unit 17 receives the meaning transmitted by the DB server 2. The question generation unit 17 generates question data in a predetermined template using the frequently occurring words received from the comment volume analysis unit 13 and the meaning received from the DB server 2. The question generation unit 17 may generate question data by using a generation AI (Artificial Intelligence). The question generation unit 17 also generates correct answer data for the question by using the meaning received from the DB server 2. The question generation unit 17 transmits the question data to the question transmission unit 18, and transmits the correct answer data and the meanings of the frequently occurring words used to generate the question data to the answer processing unit 15 (step S1162).

[0077] For example, the above-mentioned template is a character string such as "XX (frequently occurring word) is XX (meaning of the word)." The template may be changed or added depending on the organization that conducts the Web conference. To allow users to easily respond to questions during the conference, the questions are closed questions that can be answered with two options, "yes" or "no." When generating question data, the question generation unit 17 uses the meaning of another word in the word DB 23 as "XX (meaning of the word)" with a 50% probability. The correct answer data is true-false data indicating whether or not the meaning of a frequently occurring word was used as "XX (meaning of the word)."

[0078] The operation of the Web conference system 100 after step S116 will be described with reference to Fig. 10. The question sending unit 18 receives information on the user terminal 3 of the target user who sends the question from the understanding level calculation unit 16, and receives the question data from the question generation unit 17. The question sending unit 18 sends the question data to the question answering unit 34 of the user terminal 3 of the target user (step S121).

[0079] The question answering unit 34 receives the question data sent by the question sending unit 18, and displays the question and two-choice answer buttons on the display 309 (step S321). The user answers the question with "Yes" or "No." When the user presses the two-choice answer button, the question answering unit 34 generates answer data indicating the user's answer, and sends the answer data to the answer processing unit 15 of the Web conference server 1 (step S322). The answer processing unit 15 receives the answer data (step S122).

[0080] The answer processing unit 15 compares the answer data received from the question answering unit 34 with the correct answer data received from the question generating unit 17, and generates correct / incorrect data. The correct / incorrect data is true / false value data. That is, if the answer data and the correct answer data match, the correct / incorrect data indicates true, and if the answer data and the correct answer data do not match, the correct / incorrect data indicates false. The answer processing unit 15 transmits the correct / incorrect data to the understanding level calculation unit 16 (step S123).

[0081] The understanding level calculation unit 16 receives the correct / incorrect data from the answer processing unit 15, and updates the user's understanding level based on the correct / incorrect data (step S124).

[0082] An example of a process in which the understanding degree calculation unit 16 updates the understanding degree will be described with reference to Fig. 12. In this example, there are two user terminals 3, and the user of one user terminal 3 (Nichiden Goro) answers a question correctly, while the user of the other user terminal 3 (Nichiden Rokuro) answers the question incorrectly. The answer processing unit 15 receives the answer data sent from the question answering units 34 of the two user terminals 3.

[0083] The answer processing unit 15 compares the correct answer data sent by the question generation unit 17 with the answer data sent from the question answering unit 34, and generates correct / incorrect data. After that, the answer processing unit 15 sends the correct / incorrect data to the understanding level calculation unit 16. The understanding level for judgment of Goro Nichiden is, for example, 30, and the understanding level for judgment of Rokuro Nichiden is, for example, 20.

[0084] The understanding degree calculation unit 16 receives the correct / incorrect data from the answer processing unit 15. The understanding degree calculation unit 16 calculates the understanding degree by correcting the judgment understanding degree of the target user who sent the question. t is the level of understanding for judgment J t and the question correction value Q t By using this, it is expressed by the following equation (6). P t =J t +Q t (6)

[0085] If the correct / incorrect data is true, the understanding degree calculation unit 16 calculates the understanding degree for judgment J t 1 or more questions correction value Q t If the correct / incorrect data is false, the understanding degree calculation unit 16 adds the understanding degree for judgment J t In the example shown in Figure 12, since Nichiden Goro answered the question correctly, his level of understanding P t is the level of understanding for judgment J t (e.g. 30) and the question correction value Q t (For example, 10) and the sum (40). In the example shown in Figure 12, Mr. Nichiden Rokuro did not answer the question correctly, so his level of understanding P t is the level of understanding for judgment J t (e.g., 20). In this example, the question correction value Q t The default is 10, but this may be changed depending on the usage scenario.

[0086] In the above example, the degree of understanding of the user who did not answer the question correctly does not change from the degree of understanding for judgment. The understanding degree calculation unit 16 may calculate the degree of understanding of the user who did not answer the question correctly by decreasing the degree of understanding for judgment.

[0087] The operation of the Web conference system 100 after step S124 will be described with reference to Fig. 10. The understanding level calculation unit 16 accesses the understanding level DB 22 of the DB server 2. At this time, the understanding level calculation unit 16 transmits the ID received from the user identification unit 11 and the understanding level calculated in step S124 to the DB server 2 (step S125). The DB server 2 searches the understanding level DB 22 for the understanding level of the user having the ID received from the understanding level calculation unit 16, and updates the understanding level with the understanding level received from the understanding level calculation unit 16 (step S221).

[0088] The answer processing unit 15 transmits the correct / incorrect data and the meanings of the frequently occurring words to the question answering unit 34 of the user terminal 3 (step S126). The question answering unit 34 receives the correct / incorrect data and the meanings of the frequently occurring words, and displays whether the user's answer is correct or not and the meanings of the frequently occurring words on the display 309 (step S323).

[0089] The answer processing unit 15 may transmit the meaning of the frequently occurring word to the question answering unit 34 regardless of whether the user's answer is correct or not. Alternatively, the answer processing unit 15 may not transmit the meaning of the frequently occurring word to the question answering unit 34 when the user's answer is correct, and may transmit the meaning of the frequently occurring word to the question answering unit 34 when the user's answer is incorrect.

[0090] A specific example of the operation of the Web conference system 100 will be described. A Web conference consisting of multiple people is held. Each user terminal 3 transmits video data, audio data, and input data to the Web conference server 1. The Web conference server 1 calculates the understanding level for judgment based on the video data, audio data, and input data received from each user terminal 3.

[0091] The Web conference server 1 determines a threshold value based on each user's understanding level for judgment, does nothing for users whose understanding level for judgment is equal to or higher than the threshold, and asks questions about words that appear in the conference to users whose understanding level for judgment is less than the threshold. The users who are asked the questions answer with "yes" or "no." If the user's answer is correct, the Web conference server 1 calculates the user's understanding level and stores it in the understanding level DB 22 of the DB server 2.

[0092] In the following example, the Web conference server 1 calculates the understanding level for judgment, and then executes a processing loop at five-minute intervals from the start of the conference, which includes sending question data to the user terminal 3 and calculating the understanding level according to the user's answers. The following shows the flow up to 15 minutes after the start of the conference.

[0093] First, a Web conference is held with the participation of three people: a new employee, NEC A, a supervisor, NEC B, and a division manager, NEC C. Each user terminal 3 transmits video data, audio data, and input data to the Web conference server 1. The Web conference server 1 calculates a measurement value based on the video data, audio data, and input data received from each user terminal 3, and calculates a degree of understanding for judgment based on the measurement value and the attributes of each user.

[0094] Fig. 13 shows an example of the level of understanding of each user five minutes after the start of the conference and before the Web conference server 1 transmits question data to the user terminal 3. Fig. 13 shows each user's name N10, previous level of understanding P10, current measurement value M10, job title correction value C10, and judgment level of understanding J10 at this time.

[0095] When five minutes have passed since the start of the conference, the understanding degree calculation unit 16 calculates the understanding degree for judgment for the first time, so the previous understanding degree P10 is 0. t , voice detection value V t , and the non-touch detection value K t The measurement value M10 is calculated based on this.

[0096] The emotion analysis unit 12 analyzes the video data generated by the camera 31 of the user terminal 3 and calculates an emotion analysis value E t Calculate the emotion analysis value E t is the emotion analysis value at t minutes after the start of the meeting. tA The symbols in the following explanation are represented in the same manner.

[0097] Below, the emotion analysis value E of NEC A tA The emotions of NEC employee A are expressed as follows: Nihon Denki A-san: [Happiness: 10%, Trust: 5%, Fear: 2%, Surprise: 5%, Sadness: 60%, Disgust: 10%, Anger: 3%, Expectation: 5%]

[0098] Applying the above sentiment to equation (1). Positive emotion: Happiness (10%) + Trust (5%) = 15 Neutral emotion: Surprise (5%) + Anticipation (5%) = 10 Negative emotions: Sadness (60%) + Disgust (10%) + Anger (3%) + Fear (2%) = 75 Sentiment analysis value = 100 x 15 / 100 + 50 x 10 / 100 + 0 x 75 / 100 = 20

[0099] Based on the above, the sentiment analysis value E 5A becomes 20.

[0100] The speech volume analysis unit 13 analyzes the speech volume based on the voice data generated by the microphone 32 of the user terminal 3, and calculates the voice detection value V t The process of analyzing the message volume by the message volume analysis unit 13 will be described with reference to FIG.

[0101] FIG. 14 shows an example of voice data. The horizontal direction in FIG. 14 indicates time, and the vertical direction in FIG. 14 indicates the amplitude of the voice. The voice data indicates the volume of the user at each time. The total duration of the two speeches (time to unmute the microphone 32) of the NEC employee A is two minutes, and the elapsed time of the conference is five minutes. By applying these to the formula (2), the voice detection value V 5Abecomes 40(=2 / 5).

[0102] The keyboard input analysis unit 14 analyzes the frequency of input by the user based on the input data generated by the keyboard 33 of the user terminal 3, and calculates a non-keystroke detection value K t The process of analyzing the input frequency of the user by the keyboard input analysis unit 14 will be described with reference to FIG.

[0103] FIG. 15 shows an example of a user's keyboard input. The horizontal direction in FIG. 15 indicates time. The voice data indicates the characters input by the user at each time. By applying the total non-keyboard input time (2 minutes) of NEC employee A and the elapsed meeting time (5 minutes) to formula (3), the non-keyboard input detection value K 5A becomes 40(=2 / 5).

[0104] The understanding degree calculation unit 16 calculates the emotion analysis value E t , voice detection value V t , and the non-touch detection value K t is expressed as a measurement vector, and the absolute value of the measurement vector is the measurement M t Specifically, the understanding degree calculation unit 16 calculates the emotion analysis value E 5A , voice detection value V 5A , and the non-touch detection value K 5A By applying Eq. (4), the measured value M 5A The understanding level calculation unit 16 executes the same process as above for JCP B and JCP C.

[0105] The measurement value vectors and measurements of NEC A, NEC B, and NEC C are shown below. Measurement value vector of NEC A: [20, 40, 40], measurement value M 5A :60 Measurement vector of NEC B: [90, 42, 68], measurement value M 5B :120 Measurement vector of NEC C: [20,18,30], measurement value M 5C :40

[0106] The understanding level calculation unit 16 adds the position correction value C to the measured value calculated as above only the first time. The understanding level calculation unit 16 obtains the position of each user from the employee information DB 21 of the DB server 2, and determines the correction value of the position.

[0107] 16 shows an example of employee names 212 and job titles 213 in employee information held in the employee information DB 21. The understanding level calculation unit 16 acquires the job titles 213 associated with the employee names 212 of each of JEPCO employee A, JEPCO employee B, and JEPCO employee C. JEPCO employee A's job title 213 is new employee, JEPCO employee B's job title 213 is chief, and JEPCO employee C's job title 213 is division manager.

[0108] 17 shows an example of the correction value for each position. A correction value C11 is associated with a position T10 in advance. The understanding degree calculation unit 16 refers to the correction value C11 associated with each position T10 of NEC employee A, NEC employee B, and NEC employee C. The correction value C11 of NEC employee A, who is a new employee, is 5, the correction value C11 of NEC employee B, who is a manager, is 20, and the correction value C11 of NEC employee C, who is a division manager, is 200. These correction values ​​C11 are the same as the position correction value C10 shown in FIG. 13.

[0109] The understanding degree calculation unit 16 calculates the previous understanding degree P t-1 , the measured value M t , and the position correction value C are applied to equation (5) to obtain the judgment level of understanding J t Calculate the level of understanding of the three people J t is shown below. NEC A-san: Last time's understanding level P 0A (0) + current measured value M 5A (60) + Position Correction Value C A (5)=65 Nihon Denki B-san: Last time's understanding level P 0B (0) + current measured value M 5B (120) + Position Correction Value C B (20)=140 NEC C-san: Last time's understanding level P 0C (0) + current measured value M 5C (40) + Position Correction Value CC (200)=240

[0110] The understanding degree calculation unit 16 calculates the threshold value Th by multiplying the median value of the understanding degrees for judgment of all users participating in the conference. t In this example, the magnification is 0.7. Therefore, the threshold value Th t This is 98, which is 0.7 times the level of understanding (140) used for judgment by NEC employee B.

[0111] The understanding degree calculation unit 16 calculates the understanding degree for judgment J t is the threshold Th t The user who has a lower level of understanding than the threshold Th is selected as the target user to send a question. t Since the number is smaller than (98), the target user to whom the question is to be sent is JEPCO Mr. A. The Web conference server 1 does not transmit the question data to the user terminals 3 of JEPCO Mr. B and JEPCO Mr. C.

[0112] The question generation unit 17 acquires the meaning of the frequently occurring word from the word DB 23 of the DB server 2. FIG. 18 shows an example of a word 232 and a meaning 233 in the word information held by the word DB 23. When the frequently occurring word is "UcaaS", the question generation unit 17 acquires the meaning 233 associated with "UcaaS". The question generation unit 17 generates question data by using the frequently occurring word and its meaning together with a generation AI. The question generation unit 17 also generates answer data.

[0113] Examples of question data, correct answer data, and meanings of frequently occurring words are shown below. Question data: "UCaaS is a type of cloud computing service that uses virtualization technology to provide digital infrastructure such as hardware resources (CPU / memory / storage) on demand via the Internet." Correct answer: "No" Commonly used words: "UCaaS is a cloud-based solution that consolidates all of a company's business communications needs into a single solution, hosted on the cloud by a single vendor."

[0114] The question generation unit 17 transmits the question data to the question transmission unit 18, and transmits the correct answer data and the meanings of the frequently occurring words used to generate the question data to the answer processing unit 15. The question transmission unit 18 transmits the question data to the question answering unit 34 of the user terminal 3 of the target user, Mr. A of NEC Corporation. The question answering unit 34 receives the question data transmitted by the question transmission unit 18, and displays the question and answer buttons on the display 309.

[0115] 19 shows an example of a screen of the display 309 of the user terminal 3. The display 309 displays a question QU10, a "Yes" button B10, and a "No" button B11 on a screen SC10.

[0116] For example, NEC employee A presses the "No" button B11 to select "No" as the answer to the question. The question answering unit 34 transmits answer data indicating "No" to the answer processing unit 15 of the Web conference server 1. The answer processing unit 15 compares the correct answer data with the answer data and generates correct / incorrect data.

[0117] The answer processing unit 15 transmits the correct / incorrect data and the meanings of the frequently occurring words to the question answering unit 34 of the user terminal 3. The question answering unit 34 receives the correct / incorrect data and the meanings of the frequently occurring words, and displays on the display 309 whether the user's answer is correct and the meanings of the frequently occurring words.

[0118] 20 shows an example of the screen of the display 309 of the user terminal 3. The display 309 displays, on the screen SC10, a character string CH10 indicating whether the user's answer is correct or not, and a sentence TX10 indicating the meaning of the frequently occurring words. Since NEC Corporation A selects "No" as the answer, which is correct, the character string CH10 indicates that NEC Corporation A's answer is correct. By checking the sentence TX10, NEC Corporation A can deepen his understanding of the words related to the content of the meeting.

[0119] The understanding degree calculation unit 16 calculates the understanding degree for judgment J t and the question correction value Q tBy applying this to equation (6), the level of understanding P t If the user's answer is correct, the question correction value Q t is greater than or equal to 1. In this example, the question correction value Q t is 10. If the user's answer is incorrect or the question is not answered, the question correction value Q t is 0.

[0120] Since the answer of NEC's A was correct, the understanding calculation unit 16 calculates the question correction value Q t (10) is used to judge the level of understanding of Mr. A of NEC t (65). Since no questions were asked to NEC B and NEC C, the understanding degree calculation unit 16 adds the question correction value Q t (0) is used to judge the level of understanding of J for NEC B and NEC C. t Add to.

[0121] 3 people's level of understanding for judgment J t is shown below. JECA A-san: Understanding level for evaluation J 5A (65) + Question Correction Value Q 5A (10)=75 Nihon Denki B: Understanding level for evaluation J 5B (140) + Question Correction Value Q 5A (0)=140 NEC C-san: Understanding level for evaluation J 5C (240) + Question Correction Value Q 5A (0)=240

[0122] Fig. 21 shows an example of the level of understanding of each user at the point in time when 5 minutes have passed since the start of the conference and the Web conference server 1 has received each user's answer. Fig. 21 shows each user's name N10, previous level of understanding P10, current measurement value M10, job title correction value C10, judgment level of understanding J10, question correction value Q10, and level of understanding P11 at this point in time.

[0123] The understanding degree calculation unit 16 transmits the understanding degree calculated as described above to the DB server 2. The DB server 2 updates the understanding degree in the understanding degree DB 22 with the understanding degree received from the understanding degree calculation unit 16.

[0124] 22 shows an example of employee names 222 and levels of understanding 223 in the updated level of understanding information stored in the level of understanding DB 22 after 5 minutes have passed since the start of the meeting. The levels of understanding of JEPCO employee A, JEPCO employee B, and JEPCO employee C are 75, 140, and 240, respectively.

[0125] The understanding degree calculation unit 16 executes the above-mentioned process to obtain the measurement value M t For example, the measured values ​​of NEC A, NEC B, and NEC C are 90, 110, and 40, respectively.

[0126] The understanding degree calculation unit 16 calculates the previous understanding degree P t-1 , the measured value M t , and the position correction value C are applied to equation (5) to obtain the judgment level of understanding J t The understanding level calculation unit 16 calculates the position correction value C only for the first time after the start of the conference. t From the second time onwards, the position correction value C is added to the measurement value M t Do not add to.

[0127] 3 people's level of understanding for judgment J t is shown below. NEC A-san: Last time's understanding level P 5A (75) + current measurement value M 10A (90)=165 Nihon Denki B-san: Last time's understanding level P 5B (140) + current measurement value M 10B (110)=250 NEC C-san: Last time's understanding level P 5C (240) + current measurement value M 10C (40)=280

[0128] Fig. 23 shows an example of the level of understanding of each user 10 minutes after the start of the conference and before the Web conference server 1 transmits question data to the user terminal 3. Fig. 23 shows each user's name N10, previous level of understanding P11, current measurement value M11, job title correction value C11, and level of understanding for judgment J11 at this time.

[0129] The previous level of understanding P11 is the level of understanding 5 minutes after the start of the meeting. The levels of understanding P11 of NEC A, NEC B, and NEC C are 75, 140, and 240, respectively. As shown above, the measured values ​​M11 of NEC A, NEC B, and NEC C are 90, 110, and 40, respectively. The position correction value C11 is not added when calculating the level of understanding from the second time onwards. As shown above, the judgement levels of understanding J11 of NEC A, NEC B, and NEC C are 165, 250, and 280.

[0130] The understanding degree calculation unit 16 calculates the threshold value Th by multiplying the median value of the understanding degrees for judgment of all users participating in the conference. t In this example, the magnification is 0.7. Therefore, the threshold value Th t This is 175, which is 0.7 times the level of understanding (250) used for judgment by NEC student B.

[0131] The understanding degree calculation unit 16 calculates the understanding degree for judgment J t is the threshold Th t The user who has a lower level of understanding than the threshold Th is selected as the target user to send a question. t Since the number is smaller than (175), the target user to whom the question is to be sent is JEPCO Mr. A. The Web conference server 1 does not transmit the question data to the user terminals 3 of JEPCO Mr. B and JEPCO Mr. C.

[0132] The understanding degree calculation unit 16 calculates the understanding degree for judgment J t and the question correction value Q t By applying this to equation (6), the level of understanding P t Since the answer of Mr. A of NEC was correct, the understanding degree calculation unit 16 calculates the question correction value Q t (10) is used to judge the level of understanding of Mr. A of NEC t (165). Since no questions were asked to NEC B and NEC C, the understanding degree calculation unit 16 adds the question correction value Q t (0) is used to judge the level of understanding of J for NEC B and NEC C. t Add to.

[0133] 3 people's level of understanding for judgment Jt is shown below. JECA A-san: Understanding level for evaluation J 10A (165) + Question Correction Value Q 10A (10)=175 Nihon Denki B: Understanding level for evaluation J 10B (250) + Question Correction Value Q 10A (0)=250 NEC C-san: Understanding level for evaluation J 10C (280) + Question Correction Value Q 10A (0)=280

[0134] Fig. 24 shows an example of the level of understanding of each user 10 minutes after the start of the conference when the Web conference server 1 has received each user's answer. Fig. 24 shows each user's name N10, previous level of understanding P11, current measurement value M11, job title correction value C11, judgment level of understanding J11, question correction value Q11, and level of understanding P12 at this time.

[0135] The understanding degree calculation unit 16 transmits the understanding degree calculated as described above to the DB server 2. The DB server 2 updates the understanding degree in the understanding degree DB 22 with the understanding degree received from the understanding degree calculation unit 16.

[0136] 25 shows an example of employee names 222 and levels of understanding 223 in the updated level of understanding information held in the level of understanding DB 22 after 10 minutes have passed since the start of the meeting. The levels of understanding of JEPCO employee A, JEPCO employee B, and JEPCO employee C are 175, 250, and 280, respectively.

[0137] The understanding degree calculation unit 16 executes the above-mentioned process to obtain the measurement value M t For example, the measured values ​​of NEC A, NEC B, and NEC C are 100, 90, and 40, respectively.

[0138] The understanding degree calculation unit 16 calculates the previous understanding degree P t-1 , the measured value M t , and the position correction value C are applied to equation (5) to obtain the judgment level of understanding J t The understanding level calculation unit 16 calculates the position correction value C only for the first time after the start of the conference. tFrom the second time onwards, the position correction value C is added to the measurement value M t Do not add to.

[0139] 3 people's level of understanding for judgment J t is shown below. NEC A-san: Last time's understanding level P 10A (175) + current measurement value M 15A (100)=275 Nihon Denki B-san: Last time's understanding level P 10B (250) + current measurement value M 15B (90)=340 NEC C-san: Last time's understanding level P 10C (280) + current measurement value M 15C (40)=320

[0140] Fig. 26 shows an example of the level of understanding of each user 15 minutes after the start of the conference and before the Web conference server 1 transmits question data to the user terminal 3. Fig. 26 shows each user's name N10, previous level of understanding P12, current measurement value M12, job title correction value C11, and level of understanding for judgment J12 at this time.

[0141] The previous level of understanding P12 is the level of understanding 10 minutes after the start of the meeting. The levels of understanding P12 of NEC A, NEC B, and NEC C are 175, 250, and 280, respectively. As shown above, the measured values ​​M12 of NEC A, NEC B, and NEC C are 100, 90, and 40, respectively. The position correction value C11 is not added when calculating the level of understanding from the second time onwards. As shown above, the judgement levels of understanding J12 of NEC A, NEC B, and NEC C are 275, 340, and 320.

[0142] The understanding degree calculation unit 16 calculates the threshold value Th by multiplying the median value of the understanding degrees for judgment of all users participating in the conference. t In this example, the magnification is 0.7. Therefore, the threshold value Th t This is 224, which is 0.7 times the level of understanding (320) used for judgment by NEC C.

[0143] The understanding degree calculation unit 16 calculates the understanding degree for judgment J t is the threshold Th t The users whose understanding level is less than the threshold Th are selected as the target users to send questions to. t Since the number of questions is larger than (175), the Web conference server 1 does not transmit the question data to the user terminals 3 of JEPCO A, JEPCO B, and JEPCO C.

[0144] The understanding degree calculation unit 16 calculates the understanding degree for judgment J t and the question correction value Q t By applying this to equation (6), the level of understanding P t Since no questions were asked to NEC A, NEC B, and NEC C, the understanding degree calculation unit 16 calculates the question correction value Q t (0) is the level of understanding J for judging NEC A, NEC B, and NEC C. t Add to.

[0145] 3 people's level of understanding for judgment J t is shown below. JECA A-san: Understanding level for evaluation J 15A (275) + Question Correction Value Q 15A (0)=275 Nihon Denki B: Understanding level for evaluation J 15B (340) + Question Correction Value Q 15A (0)=340 NEC C-san: Understanding level for evaluation J 15C (320) + Question Correction Value Q 15A (0)=320

[0146] Fig. 27 shows an example of the level of understanding of each user when 15 minutes have passed since the start of the conference and the Web conference server 1 has received each user's answer. Fig. 27 shows each user's name N10, previous level of understanding P12, current measurement value M12, job title correction value C11, judgment level of understanding J12, question correction value Q12, and level of understanding P13 at this time.

[0147] The understanding degree calculation unit 16 transmits the understanding degree calculated as described above to the DB server 2. The DB server 2 updates the understanding degree in the understanding degree DB 22 with the understanding degree received from the understanding degree calculation unit 16.

[0148] 28 shows an example of employee names 222 and levels of understanding 223 in the updated level of understanding information held in the level of understanding DB 22 after 15 minutes have passed since the start of the meeting. The levels of understanding of JEPCO employee A, JEPCO employee B, and JEPCO employee C are 275, 340, and 320, respectively.

[0149] After this, the Web conference system 100 executes the same processes as those described above.

[0150] In the above example, the Web conference server 1 calculates the level of understanding every time a predetermined time (5 minutes) has elapsed. The Web conference server 1 may change the interval for calculating the level of understanding to any time.

[0151] The Web conference server 1 of this embodiment transmits questions to confirm understanding of words used in a Web conference to the user terminals 3 of users whose levels of understanding are lower than a preset standard. The Web conference server 1 also generates information about the words based on the users' answers and transmits the information to the user terminals 3 of users whose levels of understanding are lower than the standard. As a result, the Web conference server 1 can improve the levels of understanding of users participating in a Web conference.

[0152] The Web conference server 1 automatically asks questions about words. Therefore, even if the user is not aware of the meaning of a word, the Web conference server 1 can improve the user's understanding. In addition, the Web conference server 1 can ask questions without interrupting the progress of the conference.

[0153] The Web conference server 1 asks questions at an appropriate frequency according to the level of understanding, which reduces disruption of the Web conference caused by an excessive number of questions, compared to a system that asks questions periodically.

[0154] The Web conference server 1 calculates the degree of understanding based on the results of analyzing the data acquired from the user terminal 3, and then updates the degree of understanding based on the user's answer to the question. The Web conference server 1 can update the degree of understanding in real time according to the progress of the Web conference.

[0155] The Web conference server 1 analyzes the user's emotions based on the video data generated by the camera 31 of the user terminal 3. The Web conference server 1 can estimate the user's level of understanding based on the user's emotions.

[0156] The Web conference server 1 analyzes the amount of speech of the user based on the voice data generated by the microphone 32 of the user terminal 3. The Web conference server 1 can estimate the level of understanding based on the amount of speech of the user.

[0157] The Web conference server 1 analyzes the input frequency of the user based on the input data generated by the keyboard 33, which is an input device of the user terminal 3. The Web conference server 1 can estimate the level of understanding of the user based on the input frequency.

[0158] <Second embodiment> An embodiment of the present disclosure will be described below with reference to the drawings. FIG. 29 shows a functional configuration of a Web conference system 100a of this embodiment. The Web conference system 100a includes a Web conference server 1, a DB server 2, and a user terminal 3a. The user terminal 3 in the Web conference system 100 shown in FIG. 1 is changed to a user terminal 3a. The user terminal 3a includes a camera 31, a microphone 32, a keyboard 33, a question and answer unit 34, and a chat system 35.

[0159] The answer processing unit 15 has the following added functions. The answer processing unit 15 sends an instruction to display a "Question" button to the question answering unit 34. The answer processing unit 15 receives chat post data and an instruction to summarize chat post data from the question answering unit 34. The answer processing unit 15 interprets the chat post data received from the question answering units 34 of all user terminals 3 as the meaning of words and summarizes them. The answer processing unit 15 generates summary data and sends the summary data to the question answering unit 34. The answer processing unit 15 receives the edited summary data from the question answering unit 34 and registers it in the word DB 23.

[0160] The question and answering unit 34 has the following added functions. The question and answering unit 34 receives a “Question” button display instruction sent by the answer processing unit 15, and displays the “Question” button on the display 309. When a user presses the “Question” button, the question and answering unit 34 sends a chat start instruction to the chat system 35 to start a chat with users participating in the conference. The question and answering unit 34 receives chat post data and a chat post data summarization instruction sent by the chat system 35, and sends the chat post data and the chat post data summarization instruction to the answer processing unit 15. The question and answering unit 34 receives summary data sent by the answer processing unit 15, and sends the summary data to the chat system 35. The question and answering unit 34 receives edited summary data sent by the chat system 35, and sends the edited summary data to the answer processing unit 15.

[0161] The chat system 35 has the following functions. The chat system 35 receives a chat start instruction sent by the question and answering unit 34, and starts a chat with users participating in the conference. The chat system 35 posts text information on the chat, and sends chat post data including the posted text information to the question and answering unit 34. The chat system 35 displays a "word registration" button on the display 309. When the user presses the "word registration" button, the chat system 35 sends an instruction to summarize the chat post data to the question and answering unit 34.

[0162] Chat system 35 receives the summary data sent by question and answering unit 34, and displays the summary data on display 309. Chat system 35 displays an "Edit" button on display 309 for editing the summary data. When the user presses the "Edit" button, chat system 35 edits the summary data. Chat system 35 displays a "Register" button on display 309 for registering the edited summary data in the word DB 23 of DB server 2. When the user presses the "Register" button, chat system 35 transmits the edited summary data to question and answering unit 34.

[0163] The question generator 17 generates question data by executing the process shown in Fig. 30. Fig. 11 is changed to Fig. 30. Details of the process shown in Fig. 30 will be described.

[0164] The question generator 17 accesses the word DB 23 of the DB server 2. At this time, the question generator 17 transmits the frequently used words received from the message amount analyzer 13 to the DB server 2 (step S1161).

[0165] The DB server 2 searches the word DB 23 for the frequently occurring word received from the question generation unit 17. If the frequently occurring word is registered in the word DB 23, the DB server 2 obtains the meaning of the frequently occurring word from the word DB 23 and transmits the meaning to the question generation unit 17. If the frequently occurring word is not registered in the word DB 23, the DB server 2 transmits information to that effect to the question generation unit 17.

[0166] The question generation unit 17 determines whether or not the frequent word is registered in the word DB 23, based on the information received from the DB server 2. When the meaning of the frequent word is received from the DB server 2, the question generation unit 17 determines that the frequent word is registered in the word DB 23. When the question generation unit 17 receives information from the DB server 2 indicating that the frequent word is not registered in the word DB 23, the question generation unit 17 determines that the frequent word is not registered in the word DB 23 (step S1163).

[0167] If the frequently occurring word is registered in the word DB 23, the question generation unit 17 uses the frequently occurring word received from the comment amount analysis unit 13 and the meaning received from the DB server 2 to generate question data in a predetermined template. The question generation unit 17 also generates correct answer data for the question by using the meaning received from the DB server 2. The question generation unit 17 transmits the question data to the question sending unit 18, and transmits the correct answer data and the meaning of the frequently occurring word used to generate the question data to the answer processing unit 15 (step S1162).

[0168] If the frequently occurring word is not registered in the word DB 23, the question generation unit 17 judges whether the frequently occurring word is a general term or an in-house term. The question generation unit 17 makes this judgment using a generation AI that learns information on the Internet and learns natural language. The question generation unit 17 asks the generation AI questions that have two choices, "yes" or "no," to indicate whether the meaning of the frequently occurring word can be understood.

[0169] A question to the generation AI has a format such as, for example, "Can you tell me the meaning of the word XX? Please answer 'yes' or 'no'" (XX is a frequently occurring word). The wording used in the question may be changed to wording suitable for the generation AI. If the generation AI answers "yes", the question generation unit 17 determines that the frequently occurring word is a general term and the generation AI can understand the meaning of the frequently occurring word. If the generation AI answers "no", the question generation unit 17 determines that the frequently occurring word is an in-house term and the generation AI cannot understand the meaning of the frequently occurring word (step S1164).

[0170] If the generation AI can understand the meaning of the frequently occurring word, the generation AI notifies the question generation unit 17 of the meaning of the frequently occurring word acquired via the Internet. The question generation unit 17 generates question data and correct answer data by using the generation AI. The question generation unit 17 transmits the question data to the question transmission unit 18, and transmits the correct answer data and the meaning of the frequently occurring word to the answer processing unit 15 (step S1165).

[0171] If the generation AI cannot understand the meaning of the frequently occurring word, the question generation unit 17 generates question data in a fixed phrase as shown in the following example. The question generation unit 17 transmits the question data to the question transmission unit 18 (step S1166). The fixed phrase may be changed depending on the usage scene. Standard phrase: "Do you know the meaning of ○○ (frequently used word)?"

[0172] If the generation AI cannot understand the meaning of the frequent words, the question generation unit 17 does not generate correct answer data, and does not transmit the correct answer data and the meaning of the frequent words to the answer processing unit 15. In this case, the answer processing unit 15 does not hold the correct answer data and the meaning of the frequent words.

[0173] The question generation unit 17 may search for frequent words on the Internet without using a generation AI. When a homepage or a post on a social networking service (SNS) containing the frequent word is found, the question generation unit 17 may extract the meaning of the frequent word from the homepage or the post on the SNS. For example, the question generation unit 17 may acquire a sentence following a sentence containing a character string "What is XX (frequent word)" as the meaning of the frequent word. In this case, the question generation unit 17 may execute the same process as step S1165. When a homepage or a post on a social networking service (SNS) containing the frequent word is not found, the question generation unit 17 may execute the same process as step S1166.

[0174] The question sending unit 18 sends the question data to the question answering unit 34 of the user terminal 3. The question answering unit 34 receives the question data sent by the question sending unit 18, and displays the question and two-choice answer buttons on the display 309. If the answer processing unit 15 holds the correct answer data, the question answering unit 34 displays a question QU10, a "Yes" button B10, and a "No" button B11 on the display 309, similar to the example shown in FIG.

[0175] FIG. 31 shows an example of the screen of the display 309 of the user terminal 3 when the answer processing unit 15 does not hold correct answer data. The display 309 displays a question QU11, a "Yes" button B10, and a "No" button B11 on the screen SC10. The question QU11 is a question generated by using a fixed phrase. When the user presses the "Yes" button B10 or the "No" button B11, the question answering unit 34 transmits the answer data to the answer processing unit 15. The answer processing unit 15 transmits the answer data to the understanding degree calculation unit 16 instead of the correct / incorrect data.

[0176] The understanding degree calculation unit 16 judges the user's answer based on the answer data. When the user presses the "Yes" button B10, the understanding degree calculation unit 16 calculates the judgment understanding degree J. t 1 or more questions correction value Q t By adding t If the user presses the "No" button B11, the understanding degree calculation unit 16 calculates the understanding degree for judgment J t Without changing the level of understanding P t Use as.

[0177] Regardless of whether the user presses the "Yes" button B10 or the "No" button B11, the answer processing unit 15 sends a "Question" button display instruction to the question answering unit 34. Upon receiving the "Question" button display instruction, the question answering unit 34 displays the "Question" button on the display 309.

[0178] 32 shows an example of the screen of the display 309 of the user terminal 3 when the question and answering unit 34 receives an instruction to display the "Question" button. The display 309 displays a sentence TX11 indicating the meaning of the frequently occurring word, a "Close" button B12, and a "Question" button B13 on the screen SC10.

[0179] If the answer processing unit 15 does not hold the correct answer data, the answer processing unit 15 does not generate correct / incorrect data. Instead of transmitting the correct / incorrect data and the meanings of the frequently occurring words to the question answering unit 34, the answer processing unit 15 transmits a fixed phrase to the question answering unit 34. The question answering unit 34 displays the fixed phrase on the display 309 as the sentence TX11.

[0180] When the user presses the "Close" button B12, the text TX11, the "Close" button B12, and the "Question" button B13 disappear. Depending on the usage scene, the "Close" button B12 may not be displayed. When the user presses the "Question" button B13, the question and answer unit 34 transmits a chat start instruction to the chat system 35.

[0181] The chat system 35 receives the chat start instruction from the question and answer unit 34 and starts a chat to ask the meaning of the word to all users participating in the conference. Fig. 33 shows an example of the screen of the display 309 of the user terminal 3 after the chat system 35 starts the chat.

[0182] Chat system 35 makes a posting PS10 such as "Please tell me the meaning of XX (word)." In response to posting PS10, other users post replies related to the word (posts PS11, PS12). Chat system 35 appropriately transmits the data posted to this chat to question and answer section 34, and question and answer section 34 transmits the data to answer processing section 15 as chat posting data.

[0183] When a chat is started, the chat system 35 displays a "Register word" button B14 on the display 309. If the user who pressed the "Ask" button B13 judges that he / she has fully understood the meaning of the word from the posts by other users, he / she presses the "Register word" button B14. When the user presses the "Register word" button B14, the chat system 35 transmits an instruction to the question and answer unit 34 to summarize the chat post data.

[0184] The question and answering unit 34 receives the chat post data summarizing instruction sent by the chat system 35, and sends the chat post data summarizing instruction to the answer processing unit 15. The answer processing unit 15 receives the chat post data summarizing instruction, and uses the generation AI to interpret and summarize the chat post data of all users participating in the conference as the meanings of words. The answer processing unit 15 generates summary data and sends the summary data to the question and answering unit 34. The summary data includes words and summarized meanings of the words.

[0185] The question and answer section 34 receives the summary data sent by the answer processing section 15, and sends the summary data to the chat system 35. The chat system 35 receives the summary data. Figure 34 shows an example of a screen on the display 309 of the user terminal 3 when the chat system 35 receives the summary data.

[0186] Display 309 displays summary data AB10, an "Edit" button B15, and a "Register" button B16 on screen SC10. A user can edit summary data AB10 by pressing the "Edit" button B15. For example, a user can correct an incorrect expression or add additional information. When a user presses the "Register" button B16, chat system 35 generates edited summary data by adding edits to summary data AB10, and transmits the edited summary data to question and answer unit 34. The edited summary data includes words and summarized meanings of the words.

[0187] The question and answering unit 34 receives the edited summary data sent by the chat system 35, and transmits the edited summary data to the answer processing unit 15. The answer processing unit 15 receives the edited summary data sent by the question and answering unit 34, and registers the words and their summarized meanings in the word DB 23 of the DB server 2. This allows the Web conference system 100a to use the registered words when generating questions from the next time onwards.

[0188] A specific example of the operation of the Web conference system 100a will be described. A Web conference consisting of multiple people is held. Each user terminal 3 transmits video data, audio data, and input data to the Web conference server 1. The Web conference server 1 calculates the understanding level for judgment based on the video data, audio data, and input data received from each user terminal 3.

[0189] The Web conference server 1 determines a threshold value from each user's degree of understanding for judgment, does nothing for users whose degree of understanding for judgment is equal to or greater than the threshold, and asks questions about words that appear in the conference to users whose degree of understanding for judgment is less than the threshold. The Web conference server 1 executes the above process at five-minute intervals. Below, an example will be described in which frequently-used words are not registered in the word DB 23 five minutes and ten minutes after the start of the conference.

[0190] First, a Web conference is held with the participation of three people: a new employee, NEC A, a supervisor, NEC B, and a division manager, NEC C. Each user terminal 3 transmits video data, audio data, and input data to the Web conference server 1. The Web conference server 1 calculates a measurement value based on the video data, audio data, and input data received from each user terminal 3, and calculates a degree of understanding for judgment based on the measurement value and the attributes of each user.

[0191] Five minutes after the start of the Web conference, the understanding level for judgment of NEC employee A falls below the threshold. The question generation unit 17 executes the process shown in FIG. 30 to generate a question about "PaaS" received as a frequent word from the speech volume analysis unit 13. The question generation unit 17 accesses the word DB 23 and attempts to obtain the meaning of "PaaS". Since the meaning of "PaaS" is not registered in the word DB 23, the question generation unit 17 obtains the meaning of the word via the Internet by using a generation AI. The question generation unit 17 generates question data by using the frequent words, their meanings, and the generation AI. The question generation unit 17 also generates correct answer data.

[0192] Examples of question data, correct answer data, and meanings of frequently occurring words are shown below. Question data: "PaaS is a type of cloud computing service that uses virtualization technology to provide digital infrastructure such as hardware resources (CPU / memory / storage) on demand via the Internet." Answer data: "No" Commonly used words: "PaaS" refers to the provision of platform functions for running applications, such as virtualized application servers and databases, as a service over the Internet."

[0193] The question generation unit 17 transmits the question data to the question transmission unit 18, and transmits the correct answer data and the meanings of the frequently occurring words used to generate the question data to the answer processing unit 15. The question transmission unit 18 transmits the question data to the question answering unit 34 of the user terminal 3 of the target user, Mr. A of NEC Corporation. The question answering unit 34 receives the question data transmitted by the question transmission unit 18, and displays the question and answer buttons on the display 309.

[0194] 35 shows an example of a screen of the display 309 of the user terminal 3. The display 309 displays a question QU12, a "Yes" button B10, and a "No" button B11 on a screen SC10.

[0195] For example, NEC employee A presses the "No" button B11 to select "No" as the answer to the question. The question answering unit 34 transmits answer data indicating "No" to the answer processing unit 15 of the Web conference server 1. The answer processing unit 15 compares the correct answer data with the answer data and generates correct / incorrect data.

[0196] The answer processing unit 15 transmits the correct / incorrect data and the meanings of the frequently occurring words to the question answering unit 34 of the user terminal 3. The question answering unit 34 receives the correct / incorrect data and the meanings of the frequently occurring words, and displays on the display 309 whether the user's answer is correct and the meanings of the frequently occurring words.

[0197] The understanding degree calculation unit 16 calculates the understanding degree and transmits the understanding degree to the DB server 2. The DB server 2 updates the understanding degree in the understanding degree DB 22 with the understanding degree received from the understanding degree calculation unit 16.

[0198] Ten minutes after the start of the Web conference, the level of understanding for judgment of Mr. A of the NEC Corporation again falls below the threshold. The question generator 17 executes the process shown in FIG. 30 to generate a question about "SV95," received as a frequent word from the speech volume analyzer 13. The question generator 17 accesses the word DB 23 and attempts to obtain the meaning of "SV95." The meaning of "SV95" is not registered in the word DB 23, and "SV95" is an in-house term that does not exist on the Internet.

[0199] The question generating unit 17 does not generate answer data, but transmits question data of a fixed phrase such as "Do you know the meaning of 'SV95'?" to the question transmitting unit 18. The question transmitting unit 18 transmits the question data to the question answering unit 34 of the user terminal 3 of the target user, NEC Mr. A. The question answering unit 34 receives the question data transmitted by the question transmitting unit 18, and displays the question and answer buttons on the display 309.

[0200] For example, NEC employee A presses the "No" button to select "No" as the answer to the question. The question answering unit 34 transmits answer data indicating "No" to the answer processing unit 15 of the Web conference server 1. The answer processing unit 15 receives the answer data and transmits it to the understanding level calculation unit 16. Because the user pressed the "No" button, the understanding level calculation unit 16 updates the understanding level with the understanding level for judgment.

[0201] The answer processing unit 15 transmits an instruction to display the “question” button to the question and answering unit 34. The question and answering unit 34 receives the instruction to display the “question” button and displays the “question” button on the display 309.

[0202] When JEC employee A presses the "Ask a question" button, question and answering unit 34 sends a chat start instruction to chat system 35. Chat system 35 starts the chat, and posts PS10, "Please tell me the meaning of SV95," as shown in FIG. 33. In response to post PS10, other users post answers related to the word (posts PS11, PS12). Chat system 35 appropriately sends the data posted to this chat to question and answering unit 34, and question and answering unit 34 sends the data to answer processing unit 15 as chat posting data.

[0203] When the user A of NEC Corporation fully understands the meaning of the word, he / she presses the “Register word” button B14 shown in Fig. 33. When the user presses the “Register word” button B14, the chat system 35 transmits an instruction to the question and answer section 34 to summarize the chat posting data.

[0204] The question and answering section transmits an instruction to summarize chat posted data to the answer processing section 15. The answer processing section 15 summarizes the chat posted data to generate summary data. The answer processing section 15 transmits the summary data to the question and answering section .

[0205] The question and answer section 34 receives the summary data sent by the answer processing section 15, and sends the summary data to the chat system 35. The chat system 35 receives the summary data, and displays on the display 309 the summary data AB10, the "Edit" button B15, and the "Register" button B16 shown in FIG.

[0206] Although NEC Corporation A can edit the summary data by pressing the "Edit" button B15, he decides that editing is not necessary and presses the "Register" button B16 without editing. When NEC Corporation A presses the "Register" button B16, chat system 35 generates edited summary data based on summary data AB10 and transmits the edited summary data to question and answer section 34.

[0207] The question answering section 34 transmits the edited summary data to the answer processing section 15. The answer processing section 15 registers in the word DB 23 the words indicated by the edited summary data and the summarized meanings of the words.

[0208] 36 shows an example of words 232 and meanings 233 in the word information held by the word DB 23. A word W10 indicating "SV95" and a meaning M10 of "SV95" are newly registered.

[0209] As described above, the Web conference system 100 can build the word DB 23 according to the contents of the conference. As the number of conferences increases, data in the word DB 23 is accumulated, and the number of times that users ask questions through chat decreases. Therefore, users can smoothly understand unknown words that appear in the conference, improving the quality of the conference.

[0210] If the frequently occurring word is not registered in the word DB 23, the question generator 17 acquires the meaning of the frequently occurring word via the Internet and generates question data based on the acquired meaning. Even if the frequently occurring word is not registered in the word DB 23, if its meaning can be acquired from information on the Internet, the Web conference server 1 can ask a question without interrupting the progress of the conference.

[0211] <Third embodiment> An embodiment according to the present disclosure will be described below with reference to the drawings. FIG. 37 shows a functional configuration of an information processing device 4 of this embodiment. The information processing device 4 includes a data analysis means 41, an understanding degree calculation means 42, a question generation means 43, a question transmission means 44, an answer reception means 45, and an answer processing means 46. The data analysis means 41 receives reaction data indicating the reaction of a user participating in a Web conference from a user terminal used by the user in the Web conference, and analyzes the reaction data. The understanding degree calculation means 42 calculates the user's understanding degree based on the result of the data analysis means 41 analyzing the reaction data. The question generation means 43 generates a question for confirming the understanding of a word used in the Web conference. The question transmission means 44 transmits a question to a user terminal of a user whose understanding degree is lower than a preset standard. The answer reception means 45 receives an answer to the question from the user terminal. The answer processing means 46 generates information about the word based on the answer, and transmits the information to a user terminal of a user whose understanding degree is lower than the standard.

[0212] By being provided with the above-described configuration, the information processing device 4 can improve the level of understanding of users participating in a Web conference.

[0213] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-mentioned embodiments. Various modifications that can be understood by a person skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be appropriately combined with other embodiments.

[0214] The above-described embodiment may be used in the field of education. In recent years, online classes have become common in educational institutions such as schools and cram schools. In online classes, teachers teach students in remote locations via a Web conference system. Online classes include live distribution types in which real-time interactive exchanges are conducted, as well as on-demand types in which pre-recorded classes are distributed.

[0215] For example, the above embodiment can be used for live-streaming online classes. Live-streaming online classes are basically composed of one teacher and multiple students, and the teacher needs to pay attention to all the students. However, it is difficult for the teacher to follow up with each student without disturbing the progress of the class and under time restrictions. There are cases where students cannot keep up with the class unless they are followed up by the teacher. By applying the above embodiment to online classes, the level of understanding of the class is improved, which is beneficial for both the teacher and the students.

[0216] A part or all of the above-described embodiments may be described as, but is not limited to, the following supplementary notes.

[0217] (Appendix 1) A data analysis means for receiving reaction data indicating reactions of users participating in a Web conference from user terminals used by the users in the Web conference, and analyzing the reaction data; an understanding level calculation means for calculating a level of understanding of the user based on a result of analyzing the reaction data; A question generation means for generating questions to confirm understanding of words used in the Web conference; a question sending means for sending the question to a user terminal of a user whose understanding level is lower than a preset standard; an answer receiving means for receiving an answer to the question from the user terminal; an answer processing means for generating information about the word based on the answer and transmitting the information to the user terminal of the user whose understanding level is lower than the standard; Equipped with Information processing device.

[0218] (Appendix 2) The understanding degree calculation means updates the understanding degree based on the answer. 2. An information processing device according to claim 1.

[0219] (Appendix 3) The data analysis means receives video data generated by a camera included in the user terminal as the reaction data, and analyzes the user's emotions based on the video data. 3. The information processing device according to claim 1 or 2.

[0220] (Appendix 4) the data analysis means receives voice data generated by a microphone included in the user terminal as the reaction data, and analyzes the amount of speech of the user based on the voice data; 4. An information processing device according to any one of claims 1 to 3.

[0221] (Appendix 5) the data analysis means receives input data input by the user to the user terminal as the reaction data, and analyzes an input frequency of the user based on the input data; 5. An information processing device according to any one of claims 1 to 4.

[0222] (Appendix 6) The method further comprises: receiving voice data generated by a microphone of the user terminal; and extracting words included in the user's utterance based on the voice data; When the extracted word is registered in a database that stores words and their meanings, the question generation means generates the question based on the meaning of the word. 6. An information processing device according to any one of claims 1 to 5.

[0223] (Appendix 7) if the extracted word is not registered in the database, the question generation means acquires the meaning of the word via the Internet, generates the question based on the acquired meaning, and updates the database. 7. The information processing device according to claim 6.

[0224] (Appendix 8) When the meaning of the word cannot be obtained via the Internet, the answer processing means updates the database based on the meaning of the word obtained by the user terminal from another user in a chat. 8. The information processing device according to claim 7.

[0225] (Appendix 9) An information processing device according to any one of appendix 1 to 8; the user terminal that transmits the reaction data and the answer to the information processing device and receives the question and the information from the information processing device; Equipped with Information processing system.

[0226] (Appendix 10) Receiving reaction data indicating reactions of users participating in a Web conference from user terminals used by the users in the Web conference; Analyzing the reaction data; Calculating a level of understanding of the user based on a result of analyzing the reaction data; generating questions to verify understanding of words used in the web conference; Transmitting the question to a user terminal of a user whose level of understanding is lower than a preset standard; receiving an answer to the question from the user terminal; generating information about the word based on the answer; transmitting the information to the user terminal of the user whose understanding level is lower than the standard; An information processing method for an information processing device.

[0227] (Appendix 11) updating the level of understanding based on the answer; 11. The information processing method according to claim 10.

[0228] (Appendix 12) receiving video data generated by a camera included in the user terminal as the reaction data; Analyzing the user's emotions based on the video data; 12. The information processing method according to claim 10 or 11.

[0229] (Appendix 13) receiving voice data generated by a microphone included in the user terminal as the reaction data; Analyzing the amount of speech of the user based on the voice data; 13. An information processing method according to any one of appendix 10 to 12.

[0230] (Appendix 14) receiving input data inputted by the user to the user terminal as the reaction data; Analyzing the frequency of input by the user based on the input data; 14. An information processing method according to any one of appendix 10 to 13.

[0231] (Appendix 15) receiving voice data generated by a microphone provided in the user terminal; extracting words contained in the user's utterance based on the voice data; generating the question based on the meaning of the extracted word if the extracted word is registered in a database that stores words and their meanings; 15. An information processing method according to any one of appendix 10 to 14.

[0232] (Appendix 16) If the extracted word is not registered in the database, the meaning of the word is obtained via the Internet; generating the question based on the obtained meaning; and updating said database; 16. The information processing method according to claim 15.

[0233] (Appendix 17) if the meaning of the word cannot be obtained via the Internet, the database is updated based on the meaning of the word obtained by the user terminal from another user in a chat. 17. The information processing device according to claim 16.

[0234] (Appendix 18) A program for causing a computer to execute the information processing method according to any one of appendices 10 to 17. [Explanation of symbols]

[0235] 100,100a Web conference system 1 Web conference server 2. Database Server 3,3a User terminal 4. Information processing equipment 11 User Identification Section 12 Sentiment Analysis Department 13 Speech volume analysis section 14 Keyboard input analysis section 15 Answer processing section 16 Comprehension calculation part 17 Question generation part 18 Question Submission Department 21 Employee information DB 22 Comprehension DB 23 Word Database 31 Camera 32. Mike 33 Keyboard 34 Question and answer section 35 Chat System 41 Data Analysis Methods 42 Understanding calculation method 43 Question generation means 44 Question submission method 45 Means of receiving responses 46 Response Processing Means

Claims

1. a data analysis means for receiving reaction data indicating reactions of users participating in a Web conference from user terminals used by the users in the Web conference and analyzing the reaction data, the data analysis means receiving video data generated by a camera equipped in the user terminal as the reaction data and analyzing the emotions of the users based on the video data, or receiving audio data generated by a microphone equipped in the user terminal as the reaction data and analyzing a ratio of speaking time of the users to elapsed conference time based on the audio data, or receiving input data inputted into the user terminal by the users as the reaction data and analyzing a ratio of non-input time of the users to elapsed conference time based on the input data; an understanding degree calculation means for calculating, as the user's understanding degree, any one value or a combined value of at least two of the value indicating the emotion, the value indicating the ratio of the speech time, and the value indicating the ratio of the non-input time; A question generating means for generating questions to confirm understanding of words used in the Web conference; a question sending means for sending the question to a user terminal of a user whose understanding level is lower than a preset standard; an answer receiving means for receiving an answer to the question from the user terminal; an answer processing means for generating information indicating the meaning of the word when the answer is incorrect, and transmitting the information to the user terminal of the user whose understanding level is lower than the standard; Equipped with Information processing device.

2. the understanding degree calculation means updates the understanding degree by increasing the understanding degree when the answer is correct; The information processing device according to claim 1 .

3. The understanding degree calculation means calculates the understanding degree by adding a value corresponding to the user's job title to a composite value of at least one of the values ​​indicating the emotion, the value indicating the proportion of the speech time, and the value indicating the proportion of the non-input time.

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

4. The system further comprises a word extraction means for receiving the voice data and extracting words contained in the user's utterance based on the voice data, When the extracted word is registered in a database that stores words and their meanings, the question generation means generates the question based on the meaning of the word.

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

5. if the extracted word is not registered in the database, the question generation means acquires the meaning of the word via the Internet, generates the question based on the acquired meaning, and updates the database. The information processing device according to claim 4.

6. When the meaning of the word cannot be obtained via the Internet, the answer processing means updates the database based on the meaning of the word obtained by the user terminal from another user in a chat. The information processing device according to claim 5 .

7. An information processing device according to claim 1 or 2; the user terminal that transmits the reaction data and the answer to the information processing device and receives the question and the information from the information processing device; Equipped with Information processing system.

8. receiving reaction data indicating reactions of users participating in the Web conference from user terminals used by the users in the Web conference; receiving video data generated by a camera included in the user terminal as the reaction data, and analyzing the emotions of the user based on the video data; or receiving audio data generated by a microphone included in the user terminal as the reaction data, and analyzing a ratio of the user's speaking time to the elapsed conference time based on the audio data; or receiving input data inputted into the user terminal by the user as the reaction data, and analyzing a ratio of the user's non-input time to the elapsed conference time based on the input data; calculating, as a degree of understanding of the user, any one value or a combined value of at least two of the value indicating the emotion, the value indicating the proportion of the speech time, and the value indicating the proportion of the non-input time; generating questions to confirm understanding of words used in the Web conference; Transmitting the question to a user terminal of a user whose level of understanding is lower than a preset standard; receiving an answer to the question from the user terminal; generating information indicating the meaning of the word if the answer is incorrect; transmitting the information to the user terminal of the user whose understanding level is lower than the standard; An information processing method for an information processing device.

9. A program for causing an information processing device to execute the information processing method described in claim 8.

Images