Data processing device, data processing method and data processing system

Abstract

The purpose is to enable evaluation of each of the multiple topics included in the lecture. [Solution] The data processing device 2 of this embodiment has a lecture data acquisition unit 231 that acquires lecture data including multiple text data that indicate the content of a lecture including multiple topics and are associated with a time; an identification unit 232 that identifies a time period corresponding to each of the multiple topics based on the lecture data; a reaction data acquisition unit 233 that acquires multiple reaction data that indicate the reactions of audience members to the lecture, associated with a time; and an evaluation unit 234 that outputs evaluation results for two or more of the multiple topics based on one or more reaction data that are associated with a time corresponding to the time period for each of the multiple topics.

Description

[Technical Field]

[0001] The present invention relates to a data processing device, a data processing method, and a data processing system for evaluating the content of a lecture. [Background technology]

[0002] BACKGROUND ART Conventionally, a system capable of evaluating the content of a lecture meeting is known (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2024-19947 Summary of the Invention [Problem to be solved by the invention]

[0004] In lectures, multiple speakers often give talks on multiple topics. Conventional systems evaluate the entire lecture, but are unable to evaluate each topic in a lecture that includes multiple topics.

[0005] The present invention has been made in consideration of these points, and aims to make it possible to evaluate each of a plurality of topics included in a lecture. [Means for solving the problem]

[0006] A data processing device according to a first aspect of the present invention comprises a lecture data acquisition unit that acquires lecture data including a plurality of text data indicating the content of a lecture including a plurality of topics and associated with a time; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; a reaction data acquisition unit that acquires a plurality of reaction data indicating the reactions of viewers of the lecture to the lecture, associated with a time; and an evaluation unit that outputs evaluation results for two or more of the plurality of topics based on one or more reaction data among the plurality of reaction data associated with a time corresponding to a time period for each of the plurality of topics.

[0007] The evaluation unit may output the text data corresponding to the time associated with the one or more pieces of reaction data that contributed to a predetermined evaluation result.

[0008] The evaluation unit may output the text data corresponding to the time when the number of specified reactions indicated by the reaction data is the largest in a time period for each of the plurality of topics as the text data that contributed to the specified evaluation result.

[0009] The data processing device may further have a memory unit that stores evaluation data that associates viewer reactions with evaluations of the lecture content, and the evaluation unit may output the evaluation result based on an aggregate value of reactions for each of the multiple topics based on the one or more reaction data by referring to the evaluation data.

[0010] The evaluation unit may input a prompt to the generation AI, the prompt including an aggregated value of reactions to the topic based on the one or more reaction data and the lecture data for the topic, and output factors that interested the viewer in the topic based on the results output from the generation AI.

[0011] The evaluation unit may input a prompt to the generation AI that includes multiple key points to be discussed in the lecture and the lecture data for each of the multiple topics, and output the evaluation result indicating the appropriateness of the content for each of the multiple topics based on the results output from the generation AI.

[0012] The data processing device may further have a key item acquisition unit that acquires multiple key items to be discussed in the lecture, and the evaluation unit may evaluate a topic higher the greater the degree of relevance between each of the multiple key items and multiple words contained in the lecture data corresponding to the time period of each of the multiple topics.

[0013] The video content of the lecture may be distributed via a network to the viewers who access a specified distribution address, and the response may include ceasing access to the distribution address, and the evaluation unit may evaluate a topic higher the fewer the number of viewers who ceasing access to the distribution address during a time period for each of the multiple topics.

[0014] The reaction may include an activation in the viewer's browser that changes the video content of the lecture to the top level, and the evaluation unit may evaluate a topic higher the more activations that occur in a time period for each of the multiple topics.

[0015] The identification unit may classify the lecture data into the plurality of topics by analyzing changes in context included in the lecture data, and identify time periods corresponding to each of the plurality of topics.

[0016] A data processing method according to a second aspect of the present invention includes the steps of: acquiring lecture data, executed by a computer, including a plurality of text data indicating the content of a lecture including a plurality of topics and associated with a time; identifying a time period corresponding to each of the plurality of topics based on the lecture data; acquiring a plurality of reaction data indicating the reactions of the audience of the lecture to the lecture, associated with a time; and outputting evaluation results for two or more of the plurality of topics based on one or more reaction data among the plurality of reaction data associated with a time period corresponding to each of the plurality of topics.

[0017] A data processing system according to a third aspect of the present invention comprises a user device that creates lecture data including multiple text data that indicates the content of a lecture including multiple topics and that is associated with a time, and a data processing device that evaluates the lecture based on the lecture data, wherein the user device has a data creation unit that creates the lecture data and an output unit that outputs the evaluation results output by the data processing device, and the data processing device has a lecture data acquisition unit that acquires the lecture data from the user device, an identification unit that identifies a time period corresponding to each of the multiple topics based on the lecture data, a reaction data acquisition unit that acquires multiple reaction data that indicate the reactions of viewers of the lecture to the lecture, associated with a time, and an evaluation unit that outputs evaluation results for two or more of the multiple topics based on one or more reaction data from the multiple reaction data that are associated with a time corresponding to a time period for each of the multiple topics. [Effects of the Invention]

[0018] According to the present invention, it is possible to evaluate each of a plurality of topics included in a lecture. [Brief explanation of the drawings]

[0019] [Figure 1] FIG. 2 is a diagram for explaining an outline of the operation of the data processing system S. [Figure 2]1 is a diagram illustrating an example of the configuration of a user device 1. FIG. [Figure 3] FIG. 2 is a diagram illustrating an example of the configuration of a data processing device 2. [Figure 4] FIG. 10 is a diagram illustrating an example of lecture data. [Figure 5] FIG. 10 is a diagram illustrating an example of lecture data in which a time period for each topic is identified. [Figure 6] FIG. 10 is a diagram showing an example of reaction data. [Figure 7] FIG. 10 is a diagram illustrating an example of ranking data. [Figure 8] FIG. 10 is a diagram illustrating an example of an evaluation report. [Figure 9] FIG. 10 is a diagram showing an example of a grading report of key items. [Figure 10] 10 is a flowchart showing an example of the flow of processing executed by the data processing device 2. DETAILED DESCRIPTION OF THE INVENTION

[0020] [Data Processing System S Overview] FIG. 1 is a diagram for explaining an overview of the operation of a data processing system S. The data processing system S is a system for evaluating each topic in a lecture that includes multiple topics. The lecture is, for example, a lecture (web lecture) whose content is distributed to viewers via a network. The lecture may be distributed in real time, for example, by distributing the content of a lecture that is currently being held, or it may be distributed as a recorded version of a lecture that has already taken place. The topics may be each section of the lecture (such as the introduction, each lecture, or each event) or each topic included in a lecture by a single speaker.

[0021] In the past, organizers of lectures would compare the transcribed content of a lecture with the audience's reactions by time to determine which parts of the lecture resonated with the audience. However, this type of comparison process was cumbersome, prone to errors, and time-consuming. Another problem was that the organizer's subjective opinions could be a factor in the comparison process. Therefore, the data processing system S makes it possible to easily and objectively determine which parts of a lecture resonated with the audience.

[0022] The data processing system S comprises a user device 1 and a data processing device 2. The user device 1 is an information terminal used by an organizer UH who hosts a lecture, and is, for example, a personal computer, tablet, or smartphone. The user device 1 creates lecture data that indicates the content of the lecture. The data processing device 2 is a computer that evaluates the lecture based on the lecture data, and is, for example, a server. The information terminal 3 is an information terminal used by a viewer UA who watches the lecture, and is, for example, a personal computer, tablet, or smartphone. The data processing device 2 is capable of sending and receiving data to and from the user device 1 and the information terminal 3.

[0023] The basic operation of the data processing system S will be described below with reference to FIG. 1. User device 1 creates lecture data by transcribing the speech of speaker US recorded by microphone M used by speaker US at a lecture. The lecture data indicates the content of the lecture, which includes multiple topics, and includes multiple text data associated with time (text data indicating what was said by speaker US). User device 1 transmits the created lecture data to data processing device 2 after the lecture ends (see (1) in FIG. 1). In this way, data processing device 2 acquires the lecture data.

[0024] The data processing device 2 classifies the lecture data into multiple topics by analyzing changes in context included in the acquired lecture data, and identifies time periods corresponding to each of the multiple topics. For example, the data processing device 2 identifies time periods corresponding to the introduction, the lectures, or the events as time periods corresponding to each of the multiple topics.

[0025] The content of the lecture by the speaker US is distributed to the viewers UA via the network (see (2) in Figure 1). The viewers UA who watch the lecture react by stamping, chatting with other viewers UA, activating or deactivating the browser showing the lecture, and participating in or leaving the lecture. The information terminals 3 used by the viewers UA transmit multiple pieces of reaction data indicating the viewers UA's reactions to the lecture to the data processing device 2 as needed during the lecture (see (3) in Figure 1). As a result, the data processing device 2 acquires the multiple pieces of reaction data in association with time.

[0026] When the lecture ends, the organizer UH sends an evaluation instruction to the data processing device 2 to evaluate the lecture (see (4) in Figure 1). Upon receiving the evaluation instruction, the data processing device 2 refers to the acquired reaction data to identify reaction data associated with the time corresponding to the time period for each of the multiple topics included in the lecture data, and evaluates each of the multiple topics based on the identified reaction data. For example, the data processing device 2 determines a total score for each of the multiple topics based on the number of emotion stamps, the number of chats, the number of activations or deactivations, the number of participants or dropouts, etc.

[0027] The data processing device 2 may input the acquired lecture data into the generation AI and, based on the results output from the generation AI, identify factors of interest that interested the viewer UA in the topic. Furthermore, the data processing device 2 may identify contributing comments, which are comments that contributed to the popularity of the topic, by referring to the lecture data. For example, the data processing device 2 identifies, as contributing comments, text data indicating the content of an utterance by the speaker US that corresponds to the time associated with reaction data indicating a positive reaction (like stamp, activation, etc.) from the viewer UA.

[0028] The data processing device 2 transmits the evaluation results thus identified, which indicate the overall score, interest factors, and contributions for each of the multiple topics, to the user device 1 (see (5) in FIG. 1). This allows the organizer UH, who uses the user device 1, to check the evaluation results for each of the multiple topics.

[0029] As described above, the data processing device 2 can evaluate each of the multiple topics included in the lecture. This allows the organizer UH to easily understand the evaluation of each of the multiple topics included in the lecture. Furthermore, when evaluating multiple topics, the organizer UH's subjective opinion can be prevented from being included. The configurations and operations of the user device 1 and the data processing device 2 will be described in detail below.

[0030] [Configuration of user device 1] 2 is a diagram showing an example of the configuration of user device 1. User device 1 includes a terminal communication unit 11, an operation unit 12, a display unit 13, a storage unit 14, and a control unit 15. Control unit 15 includes a data creation unit 151, an operation reception unit 152, and an output unit 153.

[0031] The terminal communication unit 11 is a communication interface for communicating with the data processing device 2 via a communication network such as the Internet. The terminal communication unit 11 transmits the lecture data input from the data creation unit 151 to the data processing device 2. The terminal communication unit 11 also transmits an evaluation instruction input from the operation reception unit 152 to the data processing device 2 to evaluate each of the multiple topics included in the lecture. The terminal communication unit 11 also inputs the evaluation results of each of the multiple topics included in the lecture, received from the data processing device 2, to the output unit 153.

[0032] The operation unit 12 is a device that receives operations from the organizer UH, and is, for example, a keyboard, a mouse, or a touch panel. The display unit 13 is, for example, configured with a liquid crystal display or an organic EL (Electro-Luminescence) display. The display unit 13 displays various information in accordance with the control of the output unit 153.

[0033] The storage unit 14 is a storage medium including a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The storage unit 14 stores a program executed by the control unit 15. For example, the storage unit 14 stores a program that causes the control unit 15 to function as a data creation unit 151, an operation reception unit 152, and an output unit 153.

[0034] The control unit 15 is, for example, a CPU (Central Processing Unit). The control unit 15 executes a program stored in the storage unit 14, thereby functioning as a data creation unit 151, an operation reception unit 152, and an output unit 153.

[0035] Data creation unit 151 creates lecture data including multiple text data items that indicate the content of the lecture including multiple topics and are associated with a time. Data creation unit 151 creates the lecture data by, for example, transcribing the speech of speaker US recorded by microphone M used by speaker US. Data creation unit 151 inputs the created lecture data to terminal communication unit 11.

[0036] The operation receiving unit 152 receives various operations from the user. For example, the operation receiving unit 152 receives an evaluation instruction to evaluate each of multiple topics included in the lecture from the organizer UH after the lecture has ended. The operation receiving unit 152 inputs the received evaluation instruction to the terminal communication unit 11.

[0037] The output unit 153 outputs various types of information, for example, the evaluation result output by the data processing device 2. Specifically, the output unit 153 outputs the evaluation result input from the terminal communication unit 11 by displaying the evaluation result on the display unit 13.

[0038] [Configuration of data processing device 2] 3 is a diagram showing an example of the configuration of the data processing device 2. The data processing device 2 includes a device communication unit 21, a storage unit 22, and a control unit 23. The control unit 23 includes a lecture data acquisition unit 231, an identification unit 232, a reaction data acquisition unit 233, an evaluation unit 234, and a key item acquisition unit 235.

[0039] The device communication unit 21 is a communication interface for communicating with the user device 1 and the information terminal 3 via a communication network such as the Internet. The device communication unit 21 inputs the lecture data received from the user device 1 to the lecture data acquisition unit 231. The device communication unit 21 also inputs the reaction data received from the information terminal 3 to the reaction data acquisition unit 233.

[0040] The device communication unit 21 also inputs the evaluation instruction received from the user device 1 to the evaluation unit 234. The device communication unit 21 also inputs the multiple key points to be discussed in the lecture received from the user device 1 to the key point acquisition unit 235. The device communication unit 21 also transmits the evaluation results input from the evaluation unit 234 to the user device 1.

[0041] The storage unit 22 is a storage medium including a ROM, a RAM, etc. The storage unit 22 stores a program executed by the control unit 23. For example, the storage unit 22 stores an information processing program that causes the control unit 23 to function as a lecture data acquisition unit 231, an identification unit 232, a reaction data acquisition unit 233, an evaluation unit 234, and a key item acquisition unit 235. The storage unit 22 stores lecture data, reaction data, and evaluation data. Details of each data will be described later.

[0042] The control unit 23 is, for example, a CPU. The control unit 23 executes an information processing program stored in the storage unit 22, thereby functioning as a lecture data acquisition unit 231, an identification unit 232, a reaction data acquisition unit 233, an evaluation unit 234, and a key item acquisition unit 235.

[0043] The lecture data acquisition unit 231 acquires lecture data including multiple text data items that indicate the content of a lecture including multiple topics and are associated with a time. For example, after the lecture ends, the lecture data acquisition unit 231 acquires the lecture data by receiving the lecture data from the user device 1 via the device communication unit 21. The lecture data acquisition unit 231 may acquire the lecture data by receiving audio data indicating the speech of the speaker US and transcribing the speech indicated by the received audio data.

[0044] FIG. 4 is a diagram showing an example of lecture data. In the lecture data, time and text data are associated with each other. The time is the time when the speaker US spoke. The text data is data indicating a text transcript of what the speaker US said. The lecture data acquisition unit 231 inputs the acquired lecture data to the identification unit 232.

[0045] The identification unit 232 identifies a time period corresponding to each of the multiple topics in the lecture based on the lecture data, in order to have the evaluation unit 234 evaluate each of the multiple topics in the lecture. The identification unit 232 classifies the lecture data into multiple topics by, for example, analyzing changes in context included in the lecture data input from the lecture data acquisition unit 231, and identifies a time period corresponding to each of the multiple classified topics.

[0046] The identification unit 232 may identify the time periods corresponding to each of the multiple topics by accepting the time periods input by the organizer UH as the time periods corresponding to each of the multiple topics with reference to the text data of the lecture. However, it may be a heavy workload for the organizer UH to manually classify the lecture into multiple topics with reference to the text data.

[0047] Therefore, the identification unit 232 may identify the time period corresponding to each of the multiple topics using keywords that are set in advance by the organizer UH as words that appear frequently in each topic. The identification unit 232, for example, refers to the lecture data to identify the time period corresponding to text data from the first appearance of a keyword to the last appearance of a keyword as the time period corresponding to one topic, thereby identifying the time period corresponding to each of the multiple topics. A keyword is, for example, a term that succinctly represents the theme of one topic.

[0048] The identification unit 232 may also input a prompt to the generation AI, the prompt including the lecture data and an instruction statement for instructing the generation AI to identify a time period corresponding to each of the multiple topics included in the lecture, and identify a time period corresponding to each of the multiple topics based on the results output from the generation AI. In this way, the identification unit 232 classifies the lecture into multiple topics using keywords or the generation AI, thereby reducing the workload of the organizer UH.

[0049] FIG. 5 is a diagram showing an example of lecture data in which a time period for each topic has been identified. In the lecture data in which a time period for each topic has been identified, the topic, the time period, the time, and the text data are associated with each other. The topic is a section or topic of the lecture indicated by the text data corresponding to each time period, and may be input by the organizer UH or determined by the generation AI. The time period is a time period corresponding to each topic identified by the identification unit 232. The identification unit 232 stores the lecture data in which a time period for each topic has been identified in the memory unit 22.

[0050] The reaction data acquisition unit 233 acquires a plurality of pieces of reaction data indicating the reactions of the viewers UA to the lecture in association with time. For example, the reaction data acquisition unit 233 acquires the plurality of pieces of reaction data in association with time by receiving the reaction data in association with time from the information terminal 3 via the device communication unit 21 at any time during the lecture.

[0051] 6 is a diagram showing an example of reaction data. The reaction data may be data indicating the reaction of the viewer UA, and may be any of a positive reaction, a negative reaction, or a neutral reaction. Specifically, the reaction data is data indicating a reaction by an emotion stamp (like stamp, surprise stamp), a reaction by chat, a reaction by activating or deactivating a browser, or a reaction by participating in (entering) or leaving (leaving) a lecture.

[0052] Activating a browser means that the video content of the lecture is changed from a lower level to the top level in the viewer UA's browser. Deactivating a browser means that the video content of the lecture is changed from the top level to a lower level in the viewer UA's browser. Furthermore, when the video content of the lecture is distributed via a network to a viewer UA who has accessed a predetermined distribution address (e.g., an address for viewing the video content of the lecture), participating in (entering the room) the lecture means accessing the distribution address, and leaving (leaving the room) the lecture means ceasing access to the distribution address. The reaction data acquisition unit 233 stores the reaction data in the storage unit 22 in association with the time when the viewer UA made the reaction.

[0053] The evaluation unit 234 outputs evaluation results for two or more topics out of the plurality of topics based on one or more pieces of reaction data associated with times corresponding to the time periods of each of the plurality of topics. The evaluation unit 234, for example, refers to the reaction data (FIG. 6) stored in the storage unit 22 to identify one or more pieces of reaction data associated with times corresponding to the time periods of each of the plurality of topics in the lecture data (FIG. 5) stored in the storage unit 22, thereby identifying one or more pieces of reaction data corresponding to each of the plurality of topics.

[0054] The evaluation unit 234 then evaluates each of the multiple topics based on, for example, one or more pieces of identified reaction data. The evaluation unit 234 may evaluate a topic higher the fewer the number of viewer UAs that stopped accessing the distribution address during the time period for each of the multiple topics. By operating the evaluation unit 234 in this manner, the organizer UH can determine which topics were most likely to maintain the interest of the viewer UAs.

[0055] Furthermore, the evaluation unit 234 may evaluate a topic higher the more times the topic is activated in a time period. By the evaluation unit 234 operating in this manner, the organizer UH can understand the topics in which the viewer UA, who is listening to the audio of the lecture with the browser of the video content of the lecture at a lower layer (listening to the audio of the lecture while doing other work), was originally interested or became interested in by listening to the audio of the lecture.

[0056] The evaluation unit 234 may evaluate each of the multiple topics by calculating an aggregate value based on multiple types of reactions with reference to the evaluation data stored in the storage unit 22. The evaluation data associates the reactions of the viewer UAs with evaluations of the lecture content. The reactions of the viewer UAs serve as evaluation indices for evaluating each topic. The evaluation of the lecture content is an index-specific score ranging from 0 to 1, which is assigned to each topic depending on the value of the evaluation index. The index-specific score may be a value calculated by substituting the value of the evaluation index into a predetermined calculation formula. As an example, if the number of chats per minute is "20," the index-specific score corresponding to the number of chats is calculated as "0.8."

[0057] The following seven indicators can be used as evaluation criteria for each topic. 1. Average number of connections: The average number of connections. Indicates the level of sustained interest. The higher this number, the higher the score for each index. 2. Maximum Number of Connections: The maximum number of connections. This indicates the level of attention at the peak of each topic. The higher this number, the higher the score for each index. 3. Number of Exits per Minute: The number of viewer UAs who left the lecture (exited the room) per minute. Indicates the viewer UA retention rate. The smaller this number, the higher the score for each indicator will be. 4. Number of chats: The number of chats between viewer UAs. Indicates the viewer UA's active participation. The higher this number, the higher the index score will be. 5. Number of Emotion Stamps / Minute: The number of emotion stamps per minute. This indicates the emotional response of the viewer user agent. The higher this number, the higher the score for each index. 6. Number of active events per minute: The number of times the browser was activated per minute. This indicates the concentration of the viewer UA. The higher this number, the higher the score for each index. 7. Number of inactivity events / minute: The number of times the browser was inactive per minute. This indicates the tendency of viewer UA to leave. The smaller this number, the higher the score for each indicator will be.

[0058] The evaluation unit 234 may output an evaluation result based on an aggregated value of reactions for each of a plurality of topics based on one or more reaction data by referring to the evaluation data. The evaluation unit 234, for example, calculates an index-specific score (0 to 1 point) corresponding to each of the seven evaluation indexes described above for each of the plurality of topics, and calculates an overall score (0 to 7 points) by adding up the calculated seven index-specific scores. Then, for example, the evaluation unit 234 evaluates a topic more highly the higher the calculated overall score. In this way, the evaluation unit 234 outputs an evaluation result based on an aggregated value of multiple types of reactions, allowing the organizer UH to grasp the topics that have received high evaluations from an overall perspective.

[0059] Incidentally, the organizer UH may want to understand not only the evaluation of each topic but also the interest factors that the viewer UA has taken an interest in for each topic. Therefore, the evaluation unit 234 may input a prompt including an aggregate value of reactions for the topic based on one or more pieces of reaction data and the lecture data for the topic to the generation AI, and output the interest factors that the viewer UA has taken an interest in for the topic based on the results output from the generation AI.

[0060] The evaluation unit 234 inputs, for example, a prompt including scores for each of the seven indicators for the topic, lecture data for the topic, and instructions for identifying factors that made the viewer UA interested in the topic into the generation AI, and outputs interest factors that made the viewer UA interested in the topic based on the results output from the generation AI. In this way, by the evaluation unit 234 outputting interest factors that made the viewer UA interested in the topic, the organizer UH can provide feedback on the interest factors to the speaker US or select a speaker US for the next lecture based on the interest factors.

[0061] The evaluation unit 234 may output a table summarizing the overall score for each topic and the interest factor for each topic. The evaluation unit 234 outputs, for example, ranking data indicating the ranking of the overall score for each topic.

[0062] 7 is a diagram showing an example of ranking data. The ranking data associates a ranking based on an overall score with a topic, the overall score, the time period during the lecture corresponding to the topic, and an interest factor. By referring to the ranking data, the organizer UH can understand the overall score and interest factor of each topic and analyze what was good about a topic that was ranked high based on the overall score, or what was bad about a topic that was ranked low based on the overall score.

[0063] The organizer UH may also want to understand the contributing comments, which are comments that contributed to the excitement of each topic. Therefore, the evaluation unit 234 may output, as the contributing comments, text data corresponding to the time associated with one or more pieces of reaction data that contributed to a predetermined evaluation result by referring to the lecture data. The predetermined evaluation result may be, for example, a good evaluation result, or more specifically, a result in which the overall score is relatively high compared to other topics.

[0064] The evaluation unit 234 may refer to the lecture data and output, as text data that contributed to a predetermined evaluation result, text data corresponding to a time when the number of predetermined reactions indicated by the reaction data was the largest in each time period for each of the multiple topics. The predetermined reaction is, for example, a positive reaction of the viewer UA to the content of the lecture, and more specifically, pressing a "Like" stamp, posting in a chat, activating a browser, or the like. For example, by referring to the lecture data, the evaluation unit 234 outputs, as a contributing comment, text data corresponding to a second time period when the number of predetermined reactions was the largest among multiple second time periods (e.g., 10 seconds) that are shorter than the first time period and are included in the first time period corresponding to each topic.

[0065] By having the evaluation unit 234 output the contributing comments that contributed to the excitement of the topic in this way, the organizer UH can provide feedback on the contributing comments to the speaker US, ask the speaker US to prepare supplementary materials regarding the content of the contributing comments, or plan a new lecture that delves deeper into the content of the contributing comments.

[0066] The evaluation unit 234 may output the results of the analysis of the topics as described above as an evaluation report, which shows the detailed analysis results for one topic.

[0067] FIG. 8 is a diagram showing an example of an evaluation report. The evaluation report includes the topic title, the overall score, and the duration (the time period corresponding to the topic). The evaluation report also includes an area R1 showing score-related indicators, an area R2 showing interest factors, an area R3 showing the hot points of the topic (the times and types of reactions when viewer UA had relatively more positive reactions), and an area R4 showing the comments that contributed to the hot topic. By referring to the evaluation report, the organizer UH can understand the detailed analysis results for a single topic, and can accurately understand why that topic attracted the organizer UH's interest.

[0068] Incidentally, the organizer UH may want to confirm whether the topics included in the lecture have the structure or content intended by the organizer UH before the lecture. Therefore, the key item acquisition unit 235 may acquire multiple key items to be discussed in the lecture. The key items are items that the speaker US should discuss to ensure that the lecture content has the structure or content intended by the organizer UH. For example, in the case of a pharmaceutical lecture, the key items are items that are important in drug treatment (drug effectiveness, drug side effects, treatment period, risk of complications, drug cost, etc.). The key item acquisition unit 235 acquires the key items by receiving the key items specified by the organizer UH from the user device 1 via the device communication unit 21, for example, after the lecture ends.

[0069] The evaluation unit 234 may evaluate topics based on the key items by referencing the lecture data. For example, the evaluation unit 234 may evaluate a topic higher the greater the degree of association between multiple words included in the lecture data corresponding to the time period of each of the multiple topics and each of the multiple key items. The evaluation unit 234 may identify text data that matches or is similar to key words previously set by the organizer UH as words indicating the content of each of the multiple key items by referencing the lecture data corresponding to the time period of each of the multiple topics. The evaluation unit 234 may then evaluate a topic higher the greater the number of identified text data. For example, if the key item is "drug efficacy," key words include "effective," "working," "highly effective," "success," "PS (progression-free survival)," "OS (overall survival)," and "improvement."

[0070] The evaluation unit 234 may input a prompt including the key items and the lecture data for each of the multiple topics to the generation AI, and output an evaluation result indicating the appropriateness of the content for each of the multiple topics based on the results output from the generation AI. The evaluation unit 234, for example, inputs a prompt including the key items, the lecture data for each of the multiple topics, and an instruction statement for instructing the generation AI to output a score indicating the frequency of references by the speaker US to the key items for each of the multiple topics. Then, for example, based on the results output from the generation AI, the evaluation unit 234 outputs a score indicating the appropriateness of the content based on the lecture data for each of the multiple topics (hereinafter simply referred to as a "score based on the lecture data") as the evaluation result.

[0071] The evaluation unit 234 may evaluate topics based on the key items by referring to the reaction data. For example, the evaluation unit 234 may evaluate a topic higher the greater the correlation between the reaction of the viewer UA indicated by the reaction data corresponding to the time period of each of the multiple topics and each of the multiple key items. The evaluation unit 234 may identify text data that matches or is similar to the key words described above by referring to the lecture data corresponding to the time period of each of the multiple topics. Then, the evaluation unit 234 identifies reaction data associated with the time corresponding to the identified text data by referring to the reaction data, and evaluates the topic higher the greater the number of identified reaction data.

[0072] The evaluation unit 234 may input a prompt including the key points, lecture data for each of the multiple topics, and reaction data associated with the time corresponding to the time zone for each of the multiple topics to the generation AI, and output an evaluation result indicating the appropriateness of the content for each of the multiple topics based on the results output from the generation AI. For example, the evaluation unit 234 inputs a prompt including the key points, lecture data for each of the multiple topics, reaction data associated with the time corresponding to the time zone for each of the multiple topics, and an instruction statement for instructing the generation AI to output a score indicating the frequency of viewer UA reactions to the key points for each of the multiple topics. Then, for example, based on the results output from the generation AI, the evaluation unit 234 outputs a score indicating the appropriateness of the content based on the reaction data for each of the multiple topics (hereinafter simply referred to as a "score based on reaction data") as the evaluation result.

[0073] The evaluation section 234 may compare the scores based on the lecture data and the scores based on the reaction data, thus specified, to output a grading report of the key items.

[0074] 9 is a diagram showing an example of a scoring report for key points. The scoring report includes scores based on the lecture data, scores based on the reaction data, and considerations based on the scores. By referring to the scoring report for key points, the organizer UH can confirm whether the topic has the structure or content intended by the organizer UH, from the perspective of both the content of the lecture and the reaction of the audience UA.

[0075] If there is a key item that has a high score based on the lecture data but a low score based on the reaction data, the organizer UH may ask the speaker US to carefully explain the importance and significance of that key item to the audience UA in the next lecture. Also, if there is a key item that has a high score based on the reaction data but a low score based on the lecture data, the organizer UH may ask the speaker US to increase the amount of explanation of that key item in the next lecture.

[0076] Incidentally, there may be priority items that are relatively important to the organizer UH and priority items that are relatively unimportant to the organizer UH. In other words, there may be a difference in importance between the multiple priority items. Therefore, the priority item acquisition unit 235 may acquire an organizer score, which is a score indicating the degree to which the organizer UH values ​​each of the multiple priority items.

[0077] The evaluation unit 234 may then calculate, for each of the multiple key items, a difference between the organizer score and the score based on the lecture data, and a difference between the organizer score and the score based on the response data. The evaluation unit 234 may evaluate a topic higher the more key items there are for which the calculated difference is less than a threshold. By operating the evaluation unit 234 in this manner, the organizer UH can determine whether the topic has the structure or content that he or she intended before giving the lecture. As a result, the organizer UH can, for example, request the speaker US who spoke on that topic to also speak at the next lecture.

[0078] As described above, the evaluation unit 234 outputs the evaluation of the topic based on the priority items. This allows, for example, even the organizer UH who does not have the specialized knowledge to thoroughly examine the content of the topics included in the lecture to check whether the topics included in the lecture have the structure or content as intended by the organizer UH before the lecture.

[0079] [Flow of processing executed by data processing device 2] 10 is a flowchart showing an example of the flow of processing executed by the data processing device 2. After the lecture ends, the lecture data acquisition unit 231 receives the lecture data from the user device 1, thereby acquiring the lecture data (S1).

[0080] The identification unit 232 analyzes changes in context included in the lecture data acquired in S1 and classifies the lecture data into multiple topics to identify each topic (S2). The evaluation unit 234 selects an evaluation topic, which is a topic to be evaluated, from the multiple topics identified in S2 (S3). The evaluation unit 234 selects, for example, a topic corresponding to the earliest time period in the lecture data as the evaluation topic.

[0081] The evaluation unit 234 refers to the reaction data to identify one or more reaction data associated with the time corresponding to the time period of the evaluation topic selected in S3 (S4). The evaluation unit 234 identifies an overall score for the evaluation topic based on the one or more reaction data identified in S4 (S5).

[0082] The evaluation unit 234 determines whether the overall score for the evaluation topic identified in S5 is equal to or greater than a threshold (S6). The threshold may be, for example, the average, median, or minimum overall score of multiple topics that were well-received by viewers UA in past lectures.

[0083] If the evaluation unit 234 determines that the overall score of the evaluation topic is equal to or greater than the threshold (S6: YES), it refers to the lecture data and identifies, as a contributing comment, text data corresponding to the time period when the viewer UA's positive response was greatest (S7). If the evaluation unit 234 determines that the overall score of the evaluation topic is less than the threshold (S6: NO), it proceeds to step S8 without identifying a contributing comment. If the evaluation unit 234 identifies a contributing comment, it creates an evaluation report that includes the contributing comment, and if it does not identify a contributing comment, it creates an evaluation report that does not include the contributing comment.

[0084] The evaluation unit 234 determines whether evaluation of all topics identified in S2 has been completed (S8). If the evaluation unit 234 determines that evaluation of all topics has not been completed (S8: NO), the process returns to S3 and selects the topic corresponding to the next earliest time slot in the lecture data as the evaluation topic. In this way, the evaluation unit 234 repeats the processes of S3 to S8 until evaluation of all topics has been completed.

[0085] If the evaluation unit 234 determines that the evaluation of all topics has been completed (S8: YES), it outputs the evaluation results for all topics (S9). For example, the evaluation unit 234 may output a ranking of the overall scores for all topics as the evaluation result, or may output an evaluation report of a topic selected by the organizer UH from the ranking.

[0086] [Effects of Data Processing Device 2] As described above, the evaluation unit 234 outputs evaluation results for two or more of the multiple topics based on one or more pieces of reaction data associated with the time corresponding to each of the multiple topics. This allows each of the multiple topics included in the lecture to be evaluated. As a result, the organizer UH can more efficiently than ever before determine which topics resonated with the viewers UA and analyze the evaluation results for each topic in depth.

[0087] The present invention has been described above using embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist of the present invention. For example, all or part of the device can be configured by functionally or physically distributing or integrating any unit. Furthermore, new embodiments resulting from any combination of multiple embodiments are also included in the embodiments of the present invention. The effects of the new embodiments resulting from the combination also have the effects of the original embodiments. [Explanation of symbols]

[0088] 1. User Device 11. Terminal communication unit 12 Control section 13 Display section 14 Storage section 15 Control Unit 151 Data Creation Department 152 Operation reception unit 153 Output section 2. Data Processing Device 21 Device communication unit 22 Memory section 23 Control Unit 231 Lecture Data Acquisition Department 232 Specific part 233 Reaction data acquisition unit 234 Evaluation Department 235 Priority Item Acquisition Department S Data Processing System UH Organizers US Speakers UA viewers

Claims

1. a lecture data acquisition unit that acquires lecture data including a plurality of text data items that indicate the contents of a lecture including a plurality of topics and are associated with a time; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; a reaction data acquisition unit that acquires a plurality of reaction data items indicating reactions of the audience to the lecture in association with time; an evaluation unit that outputs evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and the evaluation unit outputs the text data corresponding to a time associated with the one or more pieces of reaction data that contributed to a predetermined evaluation result. Data processing device.

2. a lecture data acquisition unit that acquires lecture data including a plurality of text data items that indicate the contents of a lecture including a plurality of topics and are associated with a time; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; a reaction data acquisition unit that acquires a plurality of reaction data items indicating reactions of the audience to the lecture in association with time; an evaluation unit that outputs evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and The evaluation unit inputs a prompt including an aggregated value of reactions in the topic based on the one or more reaction data and the lecture data of the topic to a generation AI, and outputs factors that interested the viewer in the topic based on the results output from the generation AI. Data processing device.

3. a lecture data acquisition unit that acquires lecture data including a plurality of text data items that indicate the contents of a lecture including a plurality of topics and are associated with a time; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; a reaction data acquisition unit that acquires a plurality of reaction data items indicating reactions of the audience to the lecture in association with time; an evaluation unit that outputs evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and The evaluation unit inputs a prompt including a plurality of key points to be discussed in the lecture and the lecture data for each of the plurality of topics to the generation AI, and outputs the evaluation result indicating the appropriateness of the content for each of the plurality of topics based on the results output from the generation AI. Data processing device.

4. a lecture data acquisition unit that acquires lecture data including a plurality of text data items that indicate the contents of a lecture including a plurality of topics and are associated with a time; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; an important item acquisition unit that acquires a plurality of important items to be discussed in the lecture; a reaction data acquisition unit that acquires a plurality of reaction data items indicating reactions of the audience to the lecture in association with time; an evaluation unit that outputs evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and the evaluation unit evaluates the topic higher as the degree of association between each of the plurality of key points and a plurality of words included in the lecture data corresponding to a time period of each of the plurality of topics increases; Data processing device.

5. the evaluation unit outputs the text data corresponding to the time when the number of predetermined reactions indicated by the reaction data is the largest in a time period for each of the plurality of topics as the text data that contributed to the predetermined evaluation result.

2. The data processing device according to claim 1.

6. The system further includes a storage unit for storing evaluation data in which audience reactions and evaluations of the lecture content are associated with each other, the evaluation unit outputs the evaluation result based on a total value of reactions in each of the plurality of topics based on the one or more reaction data by referring to the evaluation data.

6. A data processing device according to claim 1.

7. The video content of the lecture is distributed via a network to the viewers who access a predetermined distribution address, the response includes ceasing access to the delivery address; the evaluation unit evaluates a topic higher as the number of viewers who stopped accessing the distribution address during a time period for each of the plurality of topics decreases; 6. A data processing device according to claim 1.

8. The response includes an activation in the viewer's browser that changes the video content of the lecture to the top level; the evaluation unit evaluates a topic higher as the number of activations performed in each time period of the plurality of topics increases, 6. A data processing device according to claim 1.

9. the identification unit classifies the lecture data into the plurality of topics by analyzing changes in context included in the lecture data, and identifies time periods corresponding to each of the plurality of topics.

6. A data processing device according to claim 1.

10. The computer executes acquiring lecture data including a plurality of text data items indicating contents of a lecture including a plurality of topics and associated with a time; identifying a time period corresponding to each of the plurality of topics based on the lecture data; acquiring a plurality of pieces of reaction data indicating reactions of the audience of the lecture to the lecture in association with time; outputting evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and and outputting the evaluation result, the text data corresponding to a time associated with the one or more reaction data that contributed to the predetermined evaluation result. Data processing methods.

11. The computer executes acquiring lecture data including a plurality of text data items indicating contents of a lecture including a plurality of topics and associated with a time; identifying a time period corresponding to each of the plurality of topics based on the lecture data; acquiring a plurality of pieces of reaction data indicating reactions of the audience of the lecture to the lecture in association with time; outputting evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and In the step of outputting the evaluation results, a prompt including an aggregated value of reactions in the topic based on the one or more reaction data and the lecture data of the topic is input to a generation AI, and factors that interested the viewer in the topic are output based on the results output from the generation AI. Data processing methods.

12. The computer executes acquiring lecture data including a plurality of text data items indicating contents of a lecture including a plurality of topics and associated with a time; identifying a time period corresponding to each of the plurality of topics based on the lecture data; acquiring a plurality of pieces of reaction data indicating reactions of the audience of the lecture to the lecture in association with time; outputting evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and In the step of outputting the evaluation results, a prompt including a plurality of key points to be discussed in the lecture and the lecture data for each of the plurality of topics is input to a generation AI, and the evaluation results indicating the appropriateness of the content for each of the plurality of topics are output based on the results output from the generation AI. Data processing methods.

13. The computer executes acquiring lecture data including a plurality of text data items indicating contents of a lecture including a plurality of topics and associated with a time; identifying a time period corresponding to each of the plurality of topics based on the lecture data; obtaining a plurality of key points to be discussed in the lecture; acquiring a plurality of pieces of reaction data indicating reactions of the audience of the lecture to the lecture in association with time; outputting evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and In the step of outputting the evaluation results, the higher the degree of association between each of the plurality of key points and a plurality of words included in the lecture data corresponding to the time period of each of the plurality of topics, the higher the evaluation of the topic. Data processing methods.

14. a user device that creates lecture data including a plurality of text data items that indicate the content of a lecture including a plurality of topics and are associated with a time; and a data processing device that evaluates the lecture based on the lecture data; The user device a data creation unit that creates the lecture data; an output unit that outputs the evaluation results and text data output by the data processing device; and The data processing device includes: a lecture data acquisition unit that acquires the lecture data from the user device; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; a reaction data acquisition unit that acquires a plurality of reaction data items indicating reactions of the audience to the lecture in association with time; an evaluation unit that outputs evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and the evaluation unit outputs the text data corresponding to a time associated with the one or more pieces of reaction data that contributed to a predetermined evaluation result. Data processing system.

15. a user device that creates lecture data including a plurality of text data items that indicate the content of a lecture including a plurality of topics and are associated with a time; and a data processing device that evaluates the lecture based on the lecture data; The user device a data creation unit that creates the lecture data; an output unit that outputs the evaluation results and factors of interest output by the data processing device; and The data processing device includes: a lecture data acquisition unit that acquires the lecture data from the user device; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; a reaction data acquisition unit that acquires a plurality of reaction data items indicating reactions of the audience to the lecture in association with time; an evaluation unit that outputs evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and The evaluation unit inputs a prompt including an aggregated value of reactions in the topic based on the one or more reaction data and the lecture data of the topic to the generation AI, and outputs interest factors that interest the viewer in the topic based on the results output from the generation AI. Data processing system.

16. a user device that creates lecture data including a plurality of text data items that indicate the content of a lecture including a plurality of topics and are associated with a time; and a data processing device that evaluates the lecture based on the lecture data; The user device a data creation unit that creates the lecture data; an output unit that outputs the evaluation result output by the data processing device; and The data processing device includes: a lecture data acquisition unit that acquires the lecture data from the user device; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; a reaction data acquisition unit that acquires a plurality of reaction data items indicating reactions of the audience to the lecture in association with time; an evaluation unit that outputs evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and The evaluation unit inputs a prompt including a plurality of key points to be discussed in the lecture and the lecture data for each of the plurality of topics to the generation AI, and outputs the evaluation result indicating the appropriateness of the content for each of the plurality of topics based on the results output from the generation AI. Data processing system.

17. a user device that creates lecture data including a plurality of text data items that indicate the content of a lecture including a plurality of topics and are associated with a time; and a data processing device that evaluates the lecture based on the lecture data; The user device a data creation unit that creates the lecture data; an output unit that outputs the evaluation result output by the data processing device; and The data processing device includes: a lecture data acquisition unit that acquires the lecture data from the user device; an identification unit that identifies a time period corresponding to each of the plurality of topics based on the lecture data; an important item acquisition unit that acquires a plurality of important items to be discussed in the lecture; a reaction data acquisition unit that acquires a plurality of reaction data items indicating reactions of the audience to the lecture in association with time; an evaluation unit that outputs evaluation results for two or more topics among the plurality of topics based on one or more pieces of reaction data associated with a time corresponding to a time zone for each of the plurality of topics; and the evaluation unit evaluates the topic higher as the degree of association between each of the plurality of key points and a plurality of words included in the lecture data corresponding to a time period of each of the plurality of topics increases; Data processing system.

Images