Webinar question session support device, webinar question session support method, and webinar question session support program

Abstract

It helps presenters prioritize questions and comments and assists in the appropriate order of responses. [Solution] The webinar question session support device includes a Q&A posting receiving unit that receives questions posted by viewers, a natural language analysis module that performs text analysis of the posted questions and groups similar questions, a reaction counting unit that tallys viewers' "likes" and "votes" on the questions and calculates the number of reactions for each group of grouped questions, a priority determination unit that assigns priorities to the groups of questions based on the number of grouped similar questions and the number of reactions, and a dashboard display unit that displays a list of questions organized in order of priority on a dashboard for the presenter.

Description

[Technical Field]

[0001] The present invention relates to a webinar question session support device, a webinar question session support method, and a webinar question session support program. [Background technology]

[0002] Webinars are becoming popular, where presentation data is distributed to each participant's personal computer (PC) at a remote location, and the presentation data is displayed on each participant's PC screen. In webinars with a large number of participants, it is common to use a posting tool such as Slido to conduct Q&A sessions.

[0003] For example, Patent Document 1 describes an electronic conference system that allows questions and opinions to be asked at any time even during a presentation, and that allows efficient responses to the questions and opinions. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-90350 Summary of the Invention [Problem to be solved by the invention]

[0005] The disclosures of the above-mentioned prior art documents are incorporated herein by reference. The following analysis has been carried out by the present inventors.

[0006] However, in webinars with a large number of participants, it is difficult to answer all of the questions and comments that are received, and the presenter must select the questions that they should answer on the spot. In this case, the presenter must immediately and accurately determine which questions and comments are of most interest to the audience. Therefore, it is desirable to introduce a system that shows the presenter the priority of questions and comments and supports the appropriate response order.

[0007] In view of the above-mentioned problems, the object of the present invention is to provide a webinar question session support device, a webinar question session support method, and a webinar question session support program that present the presenter with the priority of questions and comments and help support the appropriate order of responses. [Means for solving the problem]

[0008] In a first aspect of the present invention, there is provided a webinar question session support device comprising: a Q&A posting receiving unit that receives questions posted by viewers; a natural language analysis module that performs text analysis of the posted questions and groups similar questions; a reaction counting unit that counts viewers' "likes" and "votes" for the questions and calculates the number of reactions for each group of grouped questions; a priority determination unit that assigns priorities to the groups of questions based on the number of grouped similar questions and the number of reactions; and a dashboard display unit that displays a list of questions organized in order of priority on a dashboard for the presenter.

[0009] In a second aspect of the present invention, there is provided a method for supporting a webinar question session performed by an apparatus including a Q&A posting receiving unit, a natural language analysis module, a reaction counting unit, a priority determination unit, and a dashboard display unit, in which the Q&A posting receiving unit receives questions posted by viewers, the natural language analysis module performs text analysis of the posted questions and groups similar questions, the reaction counting unit counts viewers' "likes" and "votes" for the questions and calculates the number of reactions for each group of grouped questions, the priority determination unit assigns priorities to the groups of questions based on the number of similar questions grouped and the number of reactions, and the dashboard display unit displays a list of questions organized in order of priority on a dashboard for the presenter.

[0010] In a third aspect of the present invention, there is provided a webinar question session support program that causes a computer to perform the following processes: receiving questions posted by viewers; performing text analysis of the posted questions to group similar questions; tallying viewers' "likes" and "votes" for the questions and calculating the number of reactions for each group of grouped questions; assigning priorities to the question groups based on the number of grouped similar questions and the number of reactions; and displaying a list of questions organized in order of priority on a dashboard for the presenter. This program can be recorded on a computer-readable storage medium. The storage medium can be a non-transitory medium such as a semiconductor memory, a hard disk, a magnetic recording medium, or an optical recording medium. The present invention can also be embodied as a computer program product. [Effects of the Invention]

[0011] According to each aspect of the present invention, it is possible to provide a webinar question session support device, a webinar question session support method, and a webinar question session support program that present the presenter with the priority of questions and comments and help support the appropriate response order. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram illustrating a schematic configuration of a webinar question session support device according to an embodiment of the present disclosure. [Figure 2] FIG. 2 is a flowchart showing the procedure of a process performed by the webinar question session support device according to an embodiment of the present disclosure. [Figure 3] FIG. 3 is a flowchart showing a procedure of a grouping process among the processes performed by the webinar question session support device according to an embodiment of the present disclosure. [Figure 4] FIG. 4 is a diagram showing an example of a display on a dashboard used by a presenter. [Figure 5] FIG. 5 is a diagram showing an example of displaying the answer status in the question list on the viewer screen. [Figure 6] FIG. 6 is a diagram illustrating an example of the hardware configuration of the webinar question session support device. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the present invention is not limited to the embodiments described below. In addition, the same or corresponding elements in each drawing are appropriately designated by the same reference numerals. Furthermore, it should be noted that the drawings are schematic, and the dimensional relationships and ratios of each element may differ from those in reality. There may also be parts in which the dimensional relationships and ratios differ between the drawings.

[0014] 1 is a block diagram showing a schematic configuration of a webinar question session support device according to an embodiment of the present disclosure. As shown in FIG. 1, the webinar question session support device 10 processes questions and comments posted by viewers via viewer terminals 31, 32, and 33 in a webinar held using a webinar tool 20, and supports the presenter. Note that the webinar question session support device 10 can be configured as part of the terminal used by the webinar presenter, or can be configured as a server separate from the presenter's terminal.

[0015] The webinar question session support device 10 includes a Q&A posting receiving unit 11, a natural language analysis module 12, a reaction counting unit 13, a priority determining unit 14, a database 15, and a dashboard display unit 16.

[0016] The Q&A posting receiving unit 11 receives questions posted by viewers using a Q&A tool. The Q&A posting receiving unit 11 accepts questions and comments posted by viewers from terminals (for viewers) 31, 32, and 33 via a user interface such as a web application or a mobile application.

[0017] The natural language analysis module 12 performs text analysis of posted questions and comments and uses an NLP model (natural language processing model) to group similar questions.

[0018] The reaction counting unit 13 counts the number of "likes" and "votes" from viewers for each question and calculates the number of reactions for each group of grouped questions. This module counts reaction information, for example, via a Q&A tool API.

[0019] The priority determination module 14 assigns priorities to question groups based on the number of similar questions and the number of reactions (number of "likes"). This module creates a list of high-priority questions using a priority determination algorithm, which will be described later.

[0020] The database 15 stores data such as question information, reaction information, priority information, etc. The database 15 is accessed by the priority determination unit 14 and the dashboard display unit 16.

[0021] The dashboard display unit 16 displays a list of questions organized in order of priority on a dashboard for the presenter. The dashboard display unit 16 displays the list of high-priority questions in real time through a user interface such as a web application or a mobile application.

[0022] The terminals (for viewers) 31, 32, and 33 are devices (e.g., PCs, smartphones, and tablets) that viewers use to watch the Q&A session of the webinar. These terminals (for viewers) 31, 32, and 33 can post questions and tally reactions.

[0023] Next, the processing performed by the webinar question session support device 10 will be described in more detail with reference to Figures 2 and 3. Figure 2 is a flowchart showing the procedure of the processing performed by the webinar question session support device according to an embodiment of the present disclosure, and Figure 3 is a flowchart showing the procedure of the grouping processing performed by the webinar question session support device according to an embodiment of the present disclosure. Note that the following description of the processing performed by the webinar question session support device also serves as a description of the webinar question session support method according to an embodiment of the present disclosure.

[0024] As shown in FIG. 2, the processes performed by the webinar question session support device 10 include a process of receiving a question post (step S1), a process of starting text analysis of the question (step S2), a process of grouping similar questions (step S3), a process of aggregating reactions (step S4), a process of calculating priority (step S5), and a process of displaying a dashboard for presenters (step S6).

[0025] In the process of receiving a question post (step S1), the Q&A post receiving unit 11 receives a question posted by a viewer using the Q&A tool.

[0026] For example, suppose Company A is hosting a webinar to announce a new cloud service. Hundreds of viewers are participating in this webinar, and many questions are submitted during the Q&A session. Because of the large number of questions, it is not realistic to answer them all. Therefore, it is necessary to prioritize questions to be answered based on the similarity of the questions and the viewers' interest (reactions).

[0027] Viewers can "like" or "vote" for questions from other participants, and use a Q&A tool (e.g., Slido) to post questions such as the following: That is, the Q&A posting receiving unit 11 receives questions such as the following:

[0028] Participant 1: "How is this cloud service priced?" Participant 2: "Please tell me the details of your pricing plan." Participant 3: "Tell me about your security measures." Participant 4: "How long does the initial setup take?" Participant 5: "Is this service compatible with other products?" Participant 6: "What are the specific measures in terms of security?" Participant 7: “What are the customer support response times like?”

[0029] In the process of starting question text analysis (step S2), the natural language analysis module 12 starts analyzing the text of posted questions and comments. Then, in the process of grouping similar questions (step S3), the analysis results are used to group questions with similar content.

[0030] The process of grouping similar questions (step S3) will now be described with reference to Fig. 3. As shown in Fig. 3, the process of grouping similar questions includes text preprocessing (step S31), question feature extraction (step S32), question similarity calculation (step S33), similar question grouping (step S34), and group data update (step S35).

[0031] In the text preprocessing (step S31), the following preprocessing is performed on the received question.

[0032] Normalization: Removes special characters and symbols and converts all characters to lowercase or a uniform form. For example: "How much does this cloud service cost?" → "How much does this cloud service cost?"

[0033] Tokenization: Divide the question text into words or phrases. Example 1: Question 1 → Can be divided into "this", "cloud service", "price", "how", "what is it", "is it", and "is it". Example 2: Question 2 → "Price," "Plan," "Details," "Tell me," "Please"

[0034] Stopword removal: Remove common words that have no meaning. Example 1: Question 1 → "Cloud services", "Price" Example 2: Question 2 → "Price" "Plan" "Details" "Tell me"

[0035] Morphological analysis or stemming: Converting words into their root forms to improve the accuracy of the analysis.

[0036] In the question feature extraction process (step S32), a natural language processing (NLP) model is applied to quantify each question as a feature vector, which allows comparison between questions.

[0037] (1) Applying word embedding models: Mapping words into a high-dimensional vector space. Example 1: Question 1: "Price" → Vector [0.8, 0.2, -0.3, 0.1, 0.7] "Cloud services" → vector [0.6, 0.1, 0.5, 0.2, 0.4] Example 2: Question 2: "Price" → Vector [0.7, -0.1, 0.6, 0.3, 0.4] "Plan" → Vector [0.6, -0.2, 0.5, 0.4, 0.3] "Details" → Vector [0.8, 0.0, 0.7, 0.2, 0.5] "teach" → vector [0.5, -0.3, 0.4, 0.1, 0.6]

[0038] (2) Context-aware encoding: Using Sentence-BERT (similar sentence search), we encode the entire question and generate a vector that reflects the context. Example 1: Question 1 → Vector [0.45, -0.12, 0.77, ...] Example 2: Question 2 → Vector [0.42, -0.11, 0.75, ...]

[0039] In the question similarity calculation process (step S33), the cosine similarity is calculated, and the relevance between the questions is quantified as the similarity. Example 1: Similarity between Question 1 and Question 2 → 0.98 (indicating high relevance) Example 2: Similarity between Question 1 and Question 3 → 0.10 (low relevance)

[0040] In the process of grouping similar questions (step S34), questions whose similarity scores exceed a certain threshold (for example, 0.8 or more) are classified into the same group. Example of grouping results: Group 1: Price Questions Question 1: "How much does this cloud service cost?" Question 2: "Please tell me the details of your pricing plan." Group 2: Security Questions Question 3: "What security measures do you have?" Question 6: "What specific security measures are you taking?" Group 3: Initial Setup Questions Question 4: "How long does the initial setup take?" Group 4: Compatibility Questions Question 5: "Is this service compatible with other products?"

[0041] In the group data update process (step S35), when a new question is posted, the similarity with existing groups is calculated and the question is added to an appropriate group. If there is no appropriate group, a new group is created.

[0042] This concludes the description of the process of grouping similar questions (step S3). Now, returning to FIG. 2, the description of the process performed by the webinar question session support device according to the embodiment of the present disclosure will continue.

[0043] In the reaction counting process (step S4), the "likes" and "votes" given by viewers to each question are counted in real time. For example, if a group of "price questions" receives 20 "likes," it can be determined that this group has a high level of interest.

[0044] In the process of displaying the dashboard for the presenter (step S6), the list of questions organized based on priority is displayed in real time on the dashboard used by the presenter. Figure 4 shows an example of the display on the dashboard used by the presenter.

[0045] As shown in Figure 4, the presenter's dashboard displays a list of questions ranked as follows: No. 1: "Price Question" (2 similar questions, 20 likes) 2nd place: "Security Questions" (2 similar questions, 5 likes) 3rd place: "Question about initial setup" (1 like)

[0046] Finally, when the presenter answers a question, the presenter can mark the question as "answered." Once the question is "answered," the answer status is reflected in the question list on the viewer's screen. Figure 5 shows an example of how the answer status is displayed in the question list on the viewer's screen.

[0047] [Hardware configuration example] FIG. 6 is a diagram illustrating an example of the hardware configuration of a webinar question session supporting device. That is, the webinar question session supporting device 10 enables each function of the webinar question session supporting device 10 to be realized by executing the above-described webinar question session supporting method as a program on an information processing device (computer) 40 employing the hardware configuration illustrated in FIG. 6. However, the hardware configuration example illustrated in FIG. 6 is an example of a hardware configuration that realizes each function of the webinar question session supporting device 10 and is not intended to limit the hardware configuration of the webinar question session supporting device 10. The webinar question session supporting device 10 may include hardware not illustrated in FIG. 6.

[0048] As shown in FIG. 6, the hardware configuration that can be adopted by the information processing device 40 includes a CPU (Central Processing Unit) 41, a main memory device 42, an auxiliary memory device 43, and an IF (Interface) unit 44, which are interconnected by an internal bus, for example.

[0049] The CPU 41 executes each command included in the webinar question session support program executed by the information processing device (computer) 40. The main storage device 42 is, for example, a RAM (Random Access Memory), and temporarily stores various programs, such as the webinar question session support program executed by the information processing device (computer) 40, for processing by the CPU 41.

[0050] The auxiliary storage device 43 is, for example, a hard disk drive (HDD), and is capable of storing various programs, such as a webinar question session support program, executed by the information processing device (computer) 40 for a medium to long term. The various programs can be provided as program products recorded on a non-transitory computer-readable storage medium.

[0051] The IF unit 44 provides an interface for input and output with the webinar tool 20 and terminals (for viewers) 31, 32, and 33, for example.

[0052] The information processing device (computer) 40 employing the above-described hardware configuration implements each function of the webinar question session support device 10 by executing the above-described webinar question session support method as a program.

[0053] The disclosures of the above-cited patent documents and other documents are incorporated herein by reference. Modifications and adjustments of the embodiments and examples are possible within the scope of the entire disclosure of the present invention (including the claims), and further based on the basic technical concepts thereof. Furthermore, various combinations and selections (including partial deletions) of various disclosed elements (including elements of each claim, each element of each embodiment or example, each element of each drawing, etc.) are possible within the scope of the entire disclosure of the present invention. In other words, the present invention naturally embraces various modifications and alterations that would be possible by a person skilled in the art in accordance with the entire disclosure and technical concepts, including the claims. In particular, with regard to the numerical ranges set forth herein, any numerical value or subrange within that range should be construed as specifically set forth, even if not otherwise specified. Furthermore, the disclosures of the above-cited documents, when used in part or in whole in combination with the disclosures herein as part of the disclosure of the present invention, in accordance with the spirit of the present invention, are also deemed to be included in the disclosures of this application. [Explanation of symbols]

[0054] 10. Webinar Q&A Support Device 11 Q&A Submission Reception Department 12 Natural Language Analysis Module 13 Reaction counting section 14 Priority determination section 15 Databases 16 Dashboard display 20 Webinar Tools 31, 32, 33 Terminal (for viewers) 40 Information processing equipment 41 CPU 42 Main storage 43 Auxiliary storage device 44 IF Section

Claims

1. a Q&A posting receiving unit that receives questions posted by viewers; a natural language analysis module that performs text analysis of the posted questions and groups similar questions; a reaction counting unit that counts viewers' "likes" and "votes" for questions and calculates the number of reactions for each group of the grouped questions; a priority determination unit that assigns priorities to the question groups based on the number of grouped similar questions and the number of reactions; A dashboard display section for presenters that displays a list of questions organized by priority, A webinar question session support device comprising:

2. The webinar question session support device of claim 1, wherein the natural language analysis module performs normalization, tokenization, stop word removal, and morphological analysis on the text of the posted questions, quantifies the questions as feature vectors, calculates cosine similarity, quantifies the relevance between the questions as similarity, and classifies questions whose similarity score exceeds a certain threshold into the same group.

3. 3. The webinar question session support device according to claim 1, wherein the dashboard display unit displays questions answered by the presenter by marking them as "answered."

4. A webinar question session support method performed by an apparatus including a Q&A posting receiving unit, a natural language analysis module, a reaction tallying unit, a priority determining unit, and a dashboard display unit, The Q&A posting receiving unit receives questions posted by viewers, the natural language analysis module performs text analysis of the posted questions and groups similar questions; The reaction counting unit counts the viewers' "likes" and "votes" for the questions and calculates the number of reactions for each group of the grouped questions; the priority determination unit assigns priorities to the question groups based on the number of the grouped similar questions and the number of the reactions; the dashboard display unit displays a list of questions organized in order of priority on a dashboard for the presenter; How to support a webinar Q&A session.

5. 5. The webinar question session support method of claim 4, wherein the natural language analysis module performs normalization, tokenization, stop word removal, and morphological analysis on the text of the posted questions, quantifies the questions as feature vectors, calculates cosine similarity, quantifies the relevance between the questions as similarity, and classifies questions whose similarity score exceeds a certain threshold into the same group.

6. 6. The webinar question session support method according to claim 4, wherein the dashboard display unit displays questions answered by the presenter by marking them as "answered."

7. receiving questions submitted by viewers; performing text analysis of the posted questions and grouping similar questions; A process of counting viewers' "likes" and "votes" for questions and calculating the number of reactions for each group of the grouped questions; assigning priorities to the groups of questions based on the number of similar questions grouped together and the number of reactions; A dashboard for presenters showing a list of questions organized by priority, A webinar question session support program that runs on a computer.

8. 8. The webinar question session support program of claim 7, wherein the process of grouping similar questions is a process of performing normalization, tokenization, stop word removal, and morphological analysis on the text of the posted questions, quantifying the questions as feature vectors, calculating cosine similarity, quantifying the relevance between the questions as similarity, and classifying questions whose similarity score exceeds a certain threshold into the same group.

9. 9. The webinar question session support program according to claim 7, further comprising a process of marking a question answered by the presenter as "answered" and displaying the mark.

Images