Grouping method of video conference and video conference system

Abstract

The invention discloses a video conference grouping method and a video conference system. The method comprises the steps that user performance information of participants participating in a video conference is acquired, and the user performance information of the participants comprises first video information, first voice information and operation record information of the participants; and conference content information of the video conference is acquired, and the conference content information comprises second video information and second voice information. And based on a preset user attribute analysis model, determining a user attribute analysis result according to the conference participant user performance information and the conference content information. And under the condition that the video conference needs to perform grouping discussion on the participants, performing grouping processing on the participants according to the user attribute analysis result. Thus, the user attribute analysis result is determined by collecting related information and combining the preset user attribute analysis model, grouping is performed, and the member structure in each discussion group is rationalized, so that the participation degree and discussion depth of participants in each discussion group are improved.

Description

Technical Field

[0001] The present application relates to the technical field of video conferencing, and more specifically, to a video conferencing grouping method and a video conferencing system. Background Art

[0002] Given the increasing need for security and confidentiality, video conferencing applications based on quantum key distribution technology are becoming increasingly popular. In related technologies, video conferencing typically uses a random assignment mechanism to divide participants into different discussion groups during group discussions. However, this randomized grouping method can easily lead to irrational group membership, which not only affects participation enthusiasm but also reduces the overall discussion effectiveness. Summary of the Invention

[0003] The present application provides a video conferencing grouping method and a video conferencing system.

[0004] The present application provides a video conference grouping method, the method comprising:

[0005] Acquire user performance information of participants participating in the video conference, wherein the user performance information of the participants includes first video information, first voice information, and operation record information of the participants;

[0006] Acquiring conference content information of the video conference, wherein the conference content information includes second video information and second voice information;

[0007] Determining a user attribute analysis result based on a preset user attribute analysis model and the user performance information of the conference participants and the conference content information;

[0008] In the case that the video conference requires grouping of the participants for discussion, the participants are grouped according to the user attribute analysis result.

[0009] In this way, the video conferencing system obtains user performance information of participants in the video conference, where the participant user performance information includes the participant's first video information, first voice information, and operation record information. Next, the video conferencing system obtains the meeting content information of the video conference, where the meeting content information includes second video information and second voice information. Then, based on a preset user attribute analysis model, the video conferencing system determines the user attribute analysis results based on the participant user performance information and the meeting content information. Finally, in the case where the video conference requires group discussion of participants, the video conferencing system groups the participants based on the user attribute analysis results. In this way, by collecting the participant user performance information and meeting content information, combining the preset user attribute analysis model to determine the user attribute analysis results, and grouping based on the user attribute analysis results, the member structure within each discussion group is made more reasonable, thereby improving the participation and discussion depth of the participants in each discussion group.

[0010] In certain embodiments, the method further comprises:

[0011] Sending an identity authentication request and a rights management request to the terminal device used by the participant;

[0012] In a case where both the identity authentication request and the authority management request are passed, the user performance information of the conference participant and the conference content information are obtained.

[0013] In this way, the video conferencing system sends an identity authentication request and a permission management request to the terminal device used by the participant. If both the identity authentication request and the permission management request are successful, the video conferencing system obtains the participant's user performance information and meeting content information. This ensures the legitimacy of the participant's identity through the identity authentication request, preventing unauthorized users from accessing the meeting or falsifying their identities to obtain data, thereby minimizing the risk of data leakage. Furthermore, the permission management request clarifies the scope of user authorization for data collection. Only after the user's authorization is the user's performance information and meeting content information collected, protecting user privacy and enhancing user trust in the system.

[0014] In some embodiments, obtaining user performance information of participants in the video conference includes:

[0015] Receiving encrypted performance information sent by a terminal device used by the participant, wherein the encrypted performance information is obtained by encrypting the user performance information of the participant using a first quantum key obtained by the terminal device based on quantum key distribution technology;

[0016] Decrypting the encrypted performance information based on the second quantum key obtained by the quantum key distribution technology to determine the user performance information of the participant;

[0017] The obtaining of the video conference content information includes:

[0018] Receiving encrypted conference content information sent by the terminal device, wherein the encrypted conference content information is obtained by encrypting the conference content information using a third quantum key obtained by the terminal device based on the quantum key distribution technology;

[0019] The encrypted conference content information is decrypted based on the fourth quantum key obtained by the quantum key distribution technology to determine the conference content information.

[0020] In this way, the video conferencing system receives encrypted performance information sent by the terminal devices used by the participants. This encrypted performance information is generated by encrypting the participant's user performance information using a first quantum key obtained by the terminal device using quantum key distribution technology. Next, the video conferencing system decrypts the encrypted performance information using a second quantum key obtained using quantum key distribution technology to determine the participant's user performance information. Furthermore, the video conferencing system receives encrypted conference content information sent by the terminal devices. This encrypted conference content information is generated by encrypting the conference content information using a third quantum key obtained by the terminal device using quantum key distribution technology. Next, the video conferencing system decrypts the encrypted conference content information using a fourth quantum key obtained using quantum key distribution technology to determine the conference content information. In this way, the quantum key generated by quantum key distribution technology is used to encrypt the participant's user performance information and conference content information, ensuring that the data cannot be tampered with or eavesdropped during transmission, and ensuring the secure transmission of conference data from the terminal device to the cloud.

[0021] In some embodiments, the user attribute analysis result includes a first user attribute analysis result and a second user attribute analysis result, and determining the user attribute analysis result based on a preset user attribute analysis model and according to the user performance information of the conference participant and the conference content information includes:

[0022] Based on the preset user attribute analysis model and according to the user performance information of the conference participant, determining the first user attribute analysis result, the first user attribute analysis result including a speaking initiative score and a role tendency matrix;

[0023] Based on the preset user attribute analysis model, the second user attribute analysis result is determined according to the user performance information of the participant and the conference content information. The second user attribute analysis result includes a topic interest analysis result and a concept tendency analysis result.

[0024] In this way, based on the preset user attribute analysis model, the video conferencing system determines a first user attribute analysis result based on the participant's user performance information. The first user attribute analysis result includes a speaking initiative score and a role tendency matrix. Next, based on the preset user attribute analysis model, the video conferencing system determines a second user attribute analysis result based on the user's performance and meeting content information. The second user attribute analysis result includes a topic interest analysis result and a concept tendency analysis result. In this way, by using the preset user attribute analysis model to analyze the acquired participant's user performance information and meeting content information, the user attribute analysis results are determined, providing a quantitative basis for subsequent grouping and avoiding imbalanced discussions caused by single roles.

[0025] In some embodiments, grouping the participants according to the user attribute analysis results includes:

[0026] Based on preset grouping rules, grouping processing is performed according to the meeting type and the user attribute analysis results to determine the discussion group, wherein the meeting type includes a first meeting type and a second meeting type, the first meeting type is used to indicate a meeting that needs to discuss multiple topics, and the second meeting type is used to indicate a meeting that only discusses one topic.

[0027] In this way, based on preset grouping rules, the video conferencing system analyzes meeting types and user attributes to perform grouping and determine discussion groups. Meeting types include a first meeting type and a second meeting type. The first meeting type indicates a meeting that discusses multiple topics, while the second meeting type indicates a meeting that discusses only a single topic. In this way, based on preset grouping rules, different grouping methods are used for different meeting types, providing intelligent grouping services for subsequent meetings.

[0028] In some embodiments, the grouping process based on the preset grouping rules and the conference type and the user attribute analysis results to determine the discussion group includes:

[0029] In the case where the conference type is the first conference type, grouping processing is performed based on the topic interest analysis result, the concept tendency analysis result, the speech initiative score and the role tendency matrix to determine the discussion group.

[0030] Thus, for the first type of meeting, the video conferencing system groups participants based on the results of the topic interest analysis, the conceptual orientation analysis, the speaking initiative score, and the role orientation matrix to determine discussion groups. This way, when multiple topics need to be discussed, quantitative analysis of the matching degree between user attributes and the characteristics of different topics ensures a reasonable membership structure within each discussion group, enabling efficient and parallel progress of multi-topic discussions and significantly improving the overall efficiency of the meeting process.

[0031] In some embodiments, the grouping process based on the preset grouping rules and the conference type and the user attribute analysis results to determine the discussion group includes:

[0032] In the case where the meeting type is the second meeting type, grouping processing is performed based on the concept tendency analysis result, the speech initiative score and the role tendency matrix to determine the discussion group.

[0033] In this way, if the meeting type is the second type, the video conferencing system performs grouping based on the results of the conceptual analysis, the speaking activity score, and the role orientation matrix to determine the discussion groups. Thus, when discussing a single topic, by integrating multiple user attributes such as conceptual orientation, speaking activity, and role orientation, the grouping ensures a reasonable membership structure within each discussion group, significantly improving the overall efficiency of the meeting process.

[0034] In certain embodiments, the method further comprises:

[0035] When the video conference ends, the user attribute analysis result is recorded.

[0036] In this way, when a video conference ends, the video conferencing system records the user attribute analysis results. This post-conference user attribute analysis allows the video conferencing system to perform attribute analysis based on the user's long-term behavior patterns in subsequent conferences, avoiding the one-sidedness of single-conference data and improving the accuracy and adaptability of grouping.

[0037] In some embodiments, determining the user attribute analysis result based on the preset user attribute analysis model and according to the participant user performance information and the conference content information includes:

[0038] Based on the preset user attribute analysis model, the user attribute analysis result is determined according to the recorded historical user attribute analysis results, the user performance information of the conference participants and the conference content information.

[0039] Based on a pre-set user attribute analysis model, the video conferencing system determines the user attribute analysis results based on historical user attribute analysis results, participant user performance information, and meeting content information. This combination of historical user attribute analysis results avoids analysis bias caused by relying solely on data from a single video conference, ensuring that current user attribute analysis results are more closely aligned with actual user behavior patterns.

[0040] In some embodiments, the preset user attribute analysis model is trained by the following steps:

[0041] Sending a model training request to the terminal device of the participant;

[0042] If the model training request is approved, performing feature extraction processing on the user performance information of the conference participant and the conference content information to determine attribute feature information related to the user attributes;

[0043] When the video conference ends, sending a personal attribute questionnaire to the participants;

[0044] determining label data according to a personal attribute survey result obtained based on the personal attribute questionnaire;

[0045] Based on the quantum homomorphic encryption algorithm, the preset user attribute analysis model is trained according to the attribute feature information and the label data.

[0046] In this way, the video conferencing system sends a model training request to the participant's terminal device. If the model training request is approved, the video conferencing system performs feature extraction on the participant's user performance information and meeting content information to determine attribute feature information related to the user's attributes. Then, at the end of the video conference, the video conferencing system sends a personal attribute questionnaire to the participant. The video conferencing system then uses the personal attribute survey results obtained from the personal attribute questionnaire as labeled data. Finally, based on the quantum homomorphic encryption algorithm, the video conferencing system trains the preset user attribute analysis model based on the attribute feature information and labeled data. In this way, through the synergy of user authorization, feature screening, labeled data collection, and quantum encryption technology, the compliance of data collection and user privacy can be guaranteed, and the analytical and reasoning capabilities of the preset user attribute analysis model can be enhanced, thereby improving the accuracy of the output user attribute analysis results.

[0047] The embodiment of the present application provides a video conferencing system, which is used to implement the above-mentioned grouping method. The video conferencing system includes a quantum encryption module, a user attribute analysis module and a grouping module;

[0048] The quantum encryption module is configured to obtain user performance information of participants in the video conference and conference content information of the video conference;

[0049] The user attribute analysis module is configured to determine a user attribute analysis result based on a preset user attribute analysis model and according to the user performance information of the conference participants and the conference content information;

[0050] The grouping module is configured to group the participants according to the user attribute analysis result when the video conference requires grouping the participants for discussion.

[0051] In this way, the quantum encryption module can obtain user performance information of video conference participants and the meeting content information. Next, the user attribute analysis module can determine the user attribute analysis results based on the user performance information and meeting content information of the participants based on a preset user attribute analysis model. Finally, the grouping module can group participants based on the user attribute analysis results, if the video conference requires group discussion. In this way, by collecting user performance information and meeting content information of participants, combining the user attribute analysis results with the preset user attribute analysis model, and grouping based on the user attribute analysis results, the membership structure within each discussion group is more rationalized, thereby improving the participation and discussion depth of participants in each discussion group.

[0052] Additional aspects and advantages of the embodiments of the present application will be given in part in the description below, and in part will become obvious from the description below, or will be learned through practice of the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0054] Figure 1 This is one of the flow charts of the grouping method for a video conference according to an embodiment of the present application;

[0055] Figure 2 This is a second flow chart of a method for grouping a video conference according to an embodiment of the present application;

[0056] Figure 3 This is a third flow chart of a method for grouping a video conference according to an embodiment of the present application;

[0057] Figure 4 This is a fourth flow chart of a method for grouping a video conference according to an embodiment of the present application;

[0058] Figure 5 This is a fifth flowchart of a method for grouping a video conference according to an embodiment of the present application;

[0059] Figure 6 This is the sixth flow chart of the grouping method for a video conference according to an embodiment of the present application;

[0060] Figure 7 This is the seventh flow chart of the grouping method for a video conference according to an embodiment of the present application;

[0061] Figure 8 This is the eighth flow chart of the video conferencing grouping method according to the embodiment of the present application;

[0062] Figure 9 This is a ninth flowchart of a method for grouping a video conference according to an embodiment of the present application;

[0063] Figure 10 This is a tenth flowchart of a video conferencing grouping method according to an embodiment of the present application;

[0064] Figure 11 This is the eleventh flowchart of the video conferencing grouping method according to the embodiment of the present application;

[0065] Figure 12 It is a structural diagram of the video conferencing system according to the embodiment of the present application. DETAILED DESCRIPTION

[0066] The embodiments of the present application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of the present application, and should not be understood as limiting the embodiments of the present application.

[0067] Given the increasing need for security and confidentiality, video conferencing applications based on quantum key distribution (QKD) technology are becoming increasingly popular. In related technologies, video conferencing typically uses a random assignment mechanism to divide participants into different discussion groups during group discussions.

[0068] However, this randomized grouping method has significant drawbacks: when a video conference requires multiple groupings, different discussion groups need to discuss different topics, and random assignment may result in a large number of participants being assigned to discussion groups with lower interest. For example, a participant with a high interest in "Option A" may be randomly assigned to a discussion group discussing "Option A," resulting in a significant drop in their participation and insufficient depth in the discussion of the topic. Furthermore, when a video conference involves grouping participants for a single topic, random assignment cannot guarantee a reasonable balance of opinion among members within the same discussion group. This can lead to polarized opinions within the same discussion group without a moderator to coordinate, leading to meaningless arguments or awkward silences due to an imbalanced role structure (e.g., all listeners).

[0069] In addition, random allocation does not take into account the balance of speaking enthusiasm, which may easily cause some discussion groups to be overly active while others have low participation.

[0070] Based on the above questions, please refer to Figure 1 , an embodiment of the present application provides a video conference grouping method, the method comprising:

[0071] 011: Get user performance information of participants in the video conference;

[0072] 012: Get the content information of the video conference;

[0073] 013: Based on the preset user attribute analysis model, determine the user attribute analysis results according to the user performance information of the participants and the meeting content information;

[0074] 014: When a video conference requires group discussion among participants, group the participants based on the results of user attribute analysis.

[0075] The embodiment of the present application also provides a computer device, including a memory and a processor. The video conferencing grouping method of the embodiment of the present application can be implemented by the computer device of the embodiment of the present application. Specifically, a computer program is stored in the memory, and the processor is used to obtain user performance information of participants in the video conference. And obtain the conference content information of the video conference. The processor is used to determine the user attribute analysis results based on the preset user attribute analysis model, according to the user performance information of the participants and the conference content information. And when the video conference needs to group the participants for discussion, the participants are grouped according to the user attribute analysis results.

[0076] The embodiment of the present application also provides a video conference grouping device. The video conference grouping method of the embodiment of the present application can be implemented by the video conference grouping device of the embodiment of the present application. Specifically, the video conference grouping device includes an acquisition module, a determination module and a grouping module. The acquisition module is used to obtain user performance information of participants in the video conference. And obtain the conference content information of the video conference. The determination module is used to determine the user attribute analysis results based on the preset user attribute analysis model, according to the user performance information of the participants and the conference content information. The grouping module is used to group the participants according to the user attribute analysis results when the video conference needs to group the participants for discussion.

[0077] Specifically, video conferencing refers to a form of remote conferencing achieved through network technology, allowing participants in different locations to communicate and collaborate in real time through audio and video communications, data sharing, etc., breaking geographical restrictions and improving work efficiency and communication convenience. In the implementation methods of this application, video conferencing refers to quantum secure cloud conferencing, which can be implemented with the support of quantum networks, using quantum keys to encrypt online conference audio and video data streams, combined with high-intensity national secret algorithms, to ensure the security of conference data throughout the entire process of transmission, storage, encoding and decoding, encryption and decryption.

[0078] Participant user performance information includes the participant's first video information, first voice information, and operation record information. Participant user performance information is a collection of user behavior data in the meeting. Through multi-dimensional data (video, voice, operation), it comprehensively reflects the user's participation status and behavioral characteristics in the meeting, providing a data foundation for subsequent attribute analysis.

[0079] First, video information refers to video footage of meeting participants, primarily recording their facial and body language. This information is captured by the video conferencing system through the user's client camera. This information can be used to analyze non-verbal behaviors like facial expressions and body movements to help determine their reactions to the meeting content.

[0080] First, voice information refers to the voice data of conference participants, primarily used to record the user's speech content. This information is captured by the video conferencing system through the user's microphone on the client. This information can be used to analyze the user's speech content, tone, and frequency, and then assess their speaking enthusiasm, opinion orientation, and other attributes.

[0081] Operation logs refer to all user operations on the conference client, including key clicks, browsing locations, mouse movements, and keyboard input. Operation logs are behavioral data on user interactions with the conference system. They reflect user attention (e.g., browsing locations) and operational habits, and are used to supplement analysis of user interest in conference content and engagement patterns.

[0082] Meeting content information includes secondary video and audio information, specifically the video and audio data from a video conference (hereinafter referred to as the main meeting) when group discussions are not taking place. This includes the shared screen, the presenter's PowerPoint presentation, the presenter's voice, and the voices of all participants. Meeting content information can be used to identify the meeting's themes and key points, and combined with user performance information to analyze user responses to different content.

[0083] Secondary video information refers to the main meeting video footage, including video footage from the main meeting, such as the shared screen and the presenter's PowerPoint presentation. Secondary video information can be used to identify visual information within the meeting (such as PowerPoint presentation content and shared images) and assist in analyzing user interest and perspectives on different topics.

[0084] Secondary voice information refers to the main meeting audio data, specifically the audio content from the main meeting, including the speaker's voice and the voices of all participants. Secondary voice information can be used to identify discussion topics and key points within the meeting. Combined with user voice information, it can be used to analyze user engagement and perspectives on the main meeting content.

[0085] The preset user attribute analysis model refers to an artificial intelligence model obtained through training. It takes user performance information and meeting content information as input and outputs the user's personal attribute analysis results, including speaking enthusiasm, role orientation, topic interest, and concept orientation. The preset user attribute analysis model is the core of the video conferencing grouping method provided by the embodiments of this application. It uses machine learning algorithms to conduct in-depth analysis of participant user performance information and meeting content information, converting participant user performance information and meeting content information into quantifiable user attributes, providing a decision-making basis for grouping strategies.

[0086] User attribute analysis results refer to the quantitative results obtained by analyzing user performance information and meeting content information based on a preset user attribute analysis model. User attribute analysis results include the first user attribute analysis results and the second user attribute analysis results. Among them, the first user attribute analysis results refer to the speaking initiative score and the role tendency matrix. The second user attribute analysis results refer to the topic interest analysis results and the concept tendency analysis results. The user attribute analysis results are a digital expression of the behavioral characteristics of participants in the video conference. They intuitively present the participation patterns, interest preferences, and opinion tendencies of the participants through scoring and matrix forms, providing an accurate quantitative basis for subsequent intelligent grouping, and solving the problem of unreasonable member structure within the discussion group that may be caused by random allocation.

[0087] First, the user client collects user performance information of participants in the video conference in real time and sends it to the video conference system in real time.

[0088] Next, the user client collects the video conference content information in real time and sends it to the video conference system in real time.

[0089] At the same time, the video conferencing system analyzes the user performance information and conference content information of the participants in real time to determine the user attribute analysis results.

[0090] Finally, when a video conference requires group discussion among participants, the participants are grouped according to the results of user attribute analysis.

[0091] It should be noted that when a video conference requires grouping participants for discussion, users can also choose random allocation for grouping based on actual conditions.

[0092] In summary, the video conferencing system obtains user performance information of participants in the video conference, wherein the user performance information of the participants includes the first video information, the first voice information, and the operation record information of the participants. Next, the video conferencing system obtains the conference content information of the video conference, wherein the conference content information includes the second video information and the second voice information. Then, based on the preset user attribute analysis model, the video conferencing system determines the user attribute analysis results based on the user performance information and the conference content information of the participants. Finally, in the case where the video conference requires group discussion of the participants, the video conferencing system groups the participants based on the user attribute analysis results. In this way, by collecting the user performance information and conference content information of the participants, combining the preset user attribute analysis model to determine the user attribute analysis results, and grouping based on the user attribute analysis results, the member structure within each discussion group is made more reasonable, thereby improving the participation and discussion depth of the participants in each discussion group.

[0093] See also Figure 2 In certain embodiments, the method further comprises:

[0094] 015: Send identity authentication request and permission management request to the terminal devices used by participants;

[0095] 016: If both the identity authentication request and the permission management request are passed, obtain the user performance information and conference content information of the participants.

[0096] In some embodiments, the video conferencing grouping device further includes a sending module configured to send an identity authentication request and a rights management request to a terminal device used by a participant. The acquiring module is configured to acquire user performance information and conference content information of the participant if both the identity authentication request and the rights management request are successful.

[0097] In some embodiments, the processor is further configured to send an identity authentication request and a rights management request to a terminal device used by a conference participant, and obtain user performance information of the conference participant and conference content information if both the identity authentication request and the rights management request are passed.

[0098] Specifically, an identity authentication request refers to a request sent by the video conferencing system to the terminal devices used by participants to verify their legitimacy. Using technologies such as quantum digital certificates, it ensures that user identities cannot be forged, preventing unauthorized users from accessing the meeting or forging identities to obtain data. Identity authentication requests ensure the authenticity and legitimacy of participant identities, preventing unauthorized users from intruding into the conference system and ensuring secure access to conference data.

[0099] Permission management requests are requests sent by the video conferencing system to the terminal devices used by participants to obtain authorization for data collection and use. These requests clarify the scope of permission for the collection, analysis, and storage of performance information (such as video, voice, and operation records) and meeting content. By sending permission management requests, users are ensured to actively authorize data collection, ensuring compliance with privacy protection regulations and preventing the unauthorized collection of sensitive information. Users can also independently control the scope of data usage, enhancing their trust in the system.

[0100] It's important to note that both identity authentication and permission management requests are sent by the video conferencing system to the user's terminal device before the video conference begins. If the identity authentication request fails, the user cannot join the conference. If the permission management request fails, the video conferencing system cannot obtain participant user performance information or conference content information.

[0101] In this way, the video conferencing system sends an identity authentication request and a permission management request to the terminal device used by the participant. If both the identity authentication request and the permission management request are successful, the video conferencing system obtains the participant's user performance information and meeting content information. This ensures the legitimacy of the participant's identity through the identity authentication request, preventing unauthorized users from accessing the meeting or falsifying their identities to obtain data, thereby minimizing the risk of data leakage. Furthermore, the permission management request clarifies the scope of user authorization for data collection. Only after the user's authorization is the user's performance information and meeting content information collected, protecting user privacy and enhancing user trust in the system.

[0102] See also Figure 3 and Figure 4 In some embodiments, step 011 (obtaining user performance information of participants in a video conference) includes:

[0103] 0111: Receive encrypted performance information sent by the terminal devices used by participants;

[0104] 0112: Based on the second quantum key obtained by quantum key distribution technology, the encrypted performance information is decrypted to determine the user performance information of the participant;

[0105] Step 012 (obtaining the video conference content information) includes:

[0106] 0121: Receive encrypted conference content information sent by terminal devices;

[0107] 0122: Based on the fourth quantum key obtained by quantum key distribution technology, the encrypted conference content information is decrypted to determine the conference content information.

[0108] In certain embodiments, the video conferencing grouping device further includes a receiving module and a decryption module. The receiving module is configured to receive encrypted performance information sent by a terminal device used by a participant. The decryption module is configured to decrypt the encrypted performance information based on a second quantum key obtained using quantum key distribution technology to determine the participant's user performance information. The receiving module is further configured to receive encrypted conference content information sent by the terminal device. The decryption module is further configured to decrypt the encrypted conference content information based on a fourth quantum key obtained using quantum key distribution technology to determine the conference content information.

[0109] In certain embodiments, the processor is further configured to receive encrypted performance information sent from a terminal device used by a conference participant, decrypt the encrypted performance information using a second quantum key obtained using quantum key distribution technology, and determine the conference participant's user performance information. The processor is further configured to receive encrypted conference content information sent from a terminal device, decrypt the encrypted conference content information using a fourth quantum key obtained using quantum key distribution technology, and determine the conference content information.

[0110] Specifically, the encrypted performance information refers to the ciphertext data generated by encrypting the participant's user performance information using the first quantum key obtained by the participant's terminal device based on quantum key distribution technology. This quantum key encryption of the participant's user performance information ensures that the data cannot be eavesdropped or tampered with during transmission from the participant's terminal device to the video conferencing system's cloud.

[0111] Quantum key distribution (QKD) refers to a key generation and distribution technology based on quantum physics principles (such as quantum entanglement). It is used to securely generate and share encryption keys between communicating parties, ensuring that the keys are resistant to eavesdropping and tampering during the distribution process. It should be noted that for the same information, the encryption key and decryption key are the same symmetric key. That is, the key generated by QKD technology is used for both encryption and decryption of data, conforming to the basic characteristics of symmetric encryption algorithms and ensuring the consistency and efficiency of the encryption and decryption processes.

[0112] The first quantum key refers to the encryption key obtained by the terminal device through quantum key distribution technology. It is used to encrypt the user performance information of the participant and generate encrypted performance information. The first quantum key is generated by the terminal and is only used to encrypt the user performance information.

[0113] The second quantum key refers to the decryption key generated by quantum key distribution technology. It is used to decrypt encrypted performance information and obtain the original user performance information. The second quantum key is generated by the cloud server through QKD technology and negotiated with the terminal device. The first and second quantum keys form a pair of symmetric keys.

[0114] Encrypted conference content information refers to the ciphertext data obtained by encrypting the conference content information of the video conference using the third quantum key obtained by the terminal device based on quantum key distribution technology.

[0115] The third quantum key refers to the encryption key obtained by the terminal device through quantum key distribution technology, which is used to encrypt the conference content information and generate encrypted conference content information.

[0116] The fourth quantum key refers to the decryption key generated by quantum key distribution technology. It is used to decrypt encrypted conference content and obtain the original conference content. The fourth quantum key is generated by the cloud server through QKD technology and negotiated with the terminal device. The third and fourth quantum keys form a pair of symmetric keys.

[0117] It should be noted that, in some implementations, the participant user performance information and the conference content information may be encrypted and decrypted using the same pair of symmetric keys.

[0118] In this way, the video conferencing system receives encrypted performance information sent by the terminal devices used by the participants. This encrypted performance information is generated by encrypting the participant's user performance information using a first quantum key obtained by the terminal device using quantum key distribution technology. Next, the video conferencing system decrypts the encrypted performance information using a second quantum key obtained using quantum key distribution technology to determine the participant's user performance information. Furthermore, the video conferencing system receives encrypted conference content information sent by the terminal devices. This encrypted conference content information is generated by encrypting the conference content information using a third quantum key obtained by the terminal device using quantum key distribution technology. Next, the video conferencing system decrypts the encrypted conference content information using a fourth quantum key obtained using quantum key distribution technology to determine the conference content information. In this way, the quantum key generated by quantum key distribution technology is used to encrypt the participant's user performance information and conference content information, ensuring that the data cannot be tampered with or eavesdropped during transmission, and ensuring the secure transmission of conference data from the terminal device to the cloud.

[0119] See also Figure 5 In some embodiments, the user attribute analysis result includes a first user attribute analysis result and a second user attribute analysis result. Step 013 (determining the user attribute analysis result based on a preset user attribute analysis model, user performance information of conference participants, and conference content information) includes:

[0120] 0131: Based on the preset user attribute analysis model and according to the user performance information of the participant, determine the first user attribute analysis result;

[0121] 0132: Based on the preset user attribute analysis model, according to the user performance information of the participant and the meeting content information, determine the second user attribute analysis result.

[0122] In some embodiments, the determination module is further configured to determine a first user attribute analysis result based on a preset user attribute analysis model and user performance information of a conference participant, and to determine a second user attribute analysis result based on the preset user attribute analysis model and user performance and conference content information.

[0123] In some embodiments, the processor is further configured to determine a first user attribute analysis result based on a preset user attribute analysis model and user performance information of a conference participant, and to determine a second user attribute analysis result based on the preset user attribute analysis model and user performance and conference content information.

[0124] Specifically, the first user attribute analysis result refers to the quantitative result obtained based on the preset user attribute analysis model, only based on the user performance information of the participants (including the first video information, the first voice information, and the operation record information), including the speaking initiative score and the role tendency matrix.

[0125] In some embodiments, the speaking initiative score is a total score of 100 points. The user's participation initiative in the meeting is comprehensively evaluated by combining the user's speaking frequency, voice duration, tone intensity and other voice information in the first voice information, the body participation (such as nodding, gestures, etc.) in the first video information and the interactive behavior (such as active speaking button click) in the operation record information. The higher the score, the higher the participation.

[0126] In certain embodiments, a role propensity matrix scores the roles that a user tends to play in a meeting (e.g., host, presenter, listener, commentator, leisurely person, active thinker, critic, etc.) in a matrix format, with each role corresponding to a score of 0-100, with higher scores indicating a greater tendency for the user to play that role. For example, user A's role propensity matrix may be {host: 21; presenter: 23; listener: 66; commentator: 89; leisurely person: 11; active thinker: 90; critic: 35}, which indicates that user A is an active thinker and is more willing to listen and comment. User B's role propensity matrix may be {host: 55; presenter: 87; listener: 32; commentator: 21; leisurely person: 10; active thinker: 89; critic: 79}, which indicates that the user is already an active thinker and is more willing to express and criticize.

[0127] The second user attribute analysis result refers to the quantitative result obtained based on the preset user attribute analysis model, combined with the user performance information of the participants and the conference content information analysis, including the topic interest analysis result and the concept tendency analysis result.

[0128] In certain embodiments, the results of the topic interest analysis are evaluated in the form of a score, which is comprehensively evaluated through data such as the user's video expressions (such as concentration), voice keywords (such as frequent mention of related words), and operation records (such as repeated browsing of related content) when discussing specific topics. The higher the score, the higher the interest.

[0129] In certain embodiments, the opinion tendency analysis results represent quantified opinions and attitudes on various topics. For example, using support as an example (-100 to +100), positive numbers indicate support (larger values ​​indicate more support), negative numbers indicate opposition (smaller values ​​indicate more opposition), and 0 indicates neutrality. This can be determined through sentiment analysis of user speech content, body language (e.g., nodding / shaking head), and user behavior (e.g., marking opposing opinions). For example, the topic interest analysis and opinion tendency analysis results for user C may be as follows:

[0130]

[0131] Among them, the project background, personnel information and other information in the first row of the matrix are the conference topics; the scores in the second row of the matrix are the results of the topic interest analysis, with a total score of 100, and the higher the score, the greater the interest; the values ​​in the third row of the matrix are the results of the concept tendency analysis.

[0132] In this way, based on the preset user attribute analysis model, the video conferencing system determines a first user attribute analysis result based on the participant's user performance information. The first user attribute analysis result includes a speaking initiative score and a role tendency matrix. Next, based on the preset user attribute analysis model, the video conferencing system determines a second user attribute analysis result based on the user's performance and meeting content information. The second user attribute analysis result includes a topic interest analysis result and a concept tendency analysis result. In this way, by using the preset user attribute analysis model to analyze the acquired participant's user performance information and meeting content information, the user attribute analysis results are determined, providing a quantitative basis for subsequent grouping and avoiding imbalanced discussions caused by single roles.

[0133] See also Figure 6 In some embodiments, step 014 (grouping participants based on user attribute analysis results) includes:

[0134] 0141: Based on the preset grouping rules, according to the meeting type and user attribute analysis results, group processing is carried out to determine the discussion group.

[0135] In some embodiments, the determination module is further configured to perform grouping processing and determine discussion groups based on preset grouping rules and analysis results of conference types and user attributes.

[0136] In some embodiments, the processor is further configured to perform grouping processing based on preset grouping rules and analysis results of conference type and user attributes to determine a discussion group.

[0137] Specifically, meeting types include the first and second meeting types. The first meeting type indicates a meeting that discusses multiple topics. For example, a meeting might discuss multiple topics simultaneously, such as "Project Background" and "Scheme Design." The second meeting type indicates a meeting that discusses only a single topic. For example, a group discussion might focus on the feasibility of "Scheme 1."

[0138] Preset grouping rules refer to strategies pre-set in the video conferencing system for grouping participants based on the conference type and user attribute analysis results. For video conferences of the first conference type, grouping is performed based on the user's topic interest analysis results, conceptual tendency analysis results, speaking initiative scores, and role tendency matrix, ensuring that users are assigned to topic groups of interest, and that the role structure and speaking initiative within the discussion group are balanced. For video conferences of the second conference type, grouping is performed based on the user's conceptual tendency analysis results, speaking initiative scores, and role tendency matrix, optimizing the discussion efficiency of a single topic by regulating conceptual tendencies (such as designing a discussion environment for the collision of ideas or the deepening of consensus), ensuring role complementarity, and a balanced speaking initiative.

[0139] In this way, based on preset grouping rules, the video conferencing system analyzes meeting types and user attributes to perform grouping and determine discussion groups. Meeting types include a first meeting type and a second meeting type. The first meeting type indicates a meeting that discusses multiple topics, while the second meeting type indicates a meeting that discusses only a single topic. In this way, based on preset grouping rules, different grouping methods are used for different meeting types, providing intelligent grouping services for subsequent meetings.

[0140] See also Figure 7 In some embodiments, step 0141 (grouping based on preset grouping rules, according to the meeting type and user attribute analysis results, and determining the discussion group) includes:

[0141] 01411: When the meeting type is the first meeting type, grouping is performed based on the results of the topic interest analysis, the results of the concept tendency analysis, the speech initiative score and the role tendency matrix to determine the discussion group.

[0142] In some embodiments, the determination module is further used to perform grouping processing to determine the discussion group based on the topic interest analysis results, concept tendency analysis results, speech initiative score and role tendency matrix when the meeting type is the first meeting type.

[0143] In some embodiments, the processor is further configured to, when the meeting type is the first meeting type, perform grouping processing based on the topic interest analysis results, the concept tendency analysis results, the speech initiative score, and the role tendency matrix to determine the discussion group.

[0144] Specifically, in a scenario where multiple topics need to be discussed, it is necessary to group the users based on four user attribute analysis results: topic interest analysis results, concept tendency analysis results, speech initiative score, and role tendency matrix.

[0145] The application logic of the topic interest analysis results is as follows: participants are sorted by their interest in different topics from highest to lowest, and are prioritized for placement in the sub-conferences with the highest interest. For example, if user A's interest in "Option 1" is 95 points and in "Option 2" is 60 points, they will be assigned to the "Option 1" discussion group. Furthermore, when the number of discussion groups for a particular topic reaches the preset upper limit, they are sorted in descending order of interest to ensure that the average interest of group members in that topic is maximized.

[0146] The application logic of the opinion tendency analysis results is as follows: If a topic needs to be discussed from different perspectives, participants with significantly different opinion tendencies on the same topic are grouped together to promote the exchange of different perspectives and avoid "homogeneous discussions." If a topic needs to be discussed in detail, participants with similar opinion tendencies are grouped together into sub-meetings to push the discussion into in-depth technical details and improve the efficiency of solution development.

[0147] The logic behind the speech activity score is as follows: The average activity of participants in a particular discussion group is calculated (this average activity represents the overall activity of the discussion group). Membership is then adjusted to keep the difference in overall activity across discussion groups within a preset threshold (e.g., ±5 points). This prevents discussion groups from being entirely composed of highly or lowly engaged participants, ensuring balanced discussion activity across all topics.

[0148] The application logic of the role preference matrix is ​​as follows: set a standard role ratio for each topic discussion group (for example, one moderator, two presenters, one listener, and one commentator). Members are screened based on the role matrix to ensure that the roles within the group complement each other. This avoids confusion or lack of depth in the discussion process due to missing roles.

[0149] In some embodiments, the complete grouping logic may be as follows: extract all discussion topics from the conference content information (such as identifying "Plan A", "Plan B", etc. through keywords), and determine the participants' interest and ideological tendencies in each topic. Then, preliminarily group the participants according to the topics of greatest interest to them, and generate an initial group list. Then, for each discussion group, check whether the distribution of ideological tendencies of the members in the discussion group meets the discussion objectives (such as whether a collision of opinions is needed), calculate the role ratio and the mean speaking enthusiasm of the members in the discussion group, and dynamically optimize by exchanging members (such as transferring redundant highly motivated speakers in a discussion group A to another discussion group B, where discussion group B is a topic that highly motivated speakers are more interested in) to determine the final discussion group.

[0150] It should be noted that in actual application scenarios, participants may not be evenly divided, all participants may not be very interested in a particular topic, most participants may have similar viewpoints, most participants may have low speaking enthusiasm scores, or most participants may have similar role preferences. In these cases, the video conferencing system will generate a grouping that the system considers relatively reasonable. If the video conference host deems this grouping imperfect, they can adjust the grouping or use random grouping to confirm a new grouping.

[0151] Thus, for the first type of meeting, the video conferencing system groups participants based on the results of the topic interest analysis, the conceptual orientation analysis, the speaking initiative score, and the role orientation matrix to determine discussion groups. This way, when multiple topics need to be discussed, quantitative analysis of the matching degree between user attributes and the characteristics of different topics ensures a reasonable membership structure within each discussion group, enabling efficient and parallel progress of multi-topic discussions and significantly improving the overall efficiency of the meeting process.

[0152] See also Figure 8 In some embodiments, step 0141 (grouping based on preset grouping rules, according to the meeting type and user attribute analysis results, and determining the discussion group) includes:

[0153] 01412: When the meeting type is the second meeting type, grouping is carried out according to the results of the concept tendency analysis, the speaking initiative score and the role tendency matrix to determine the discussion group.

[0154] In some embodiments, the determination module is further configured to, when the meeting type is the second meeting type, perform grouping processing based on the concept tendency analysis results, the speech initiative score, and the role tendency matrix to determine the discussion group.

[0155] In some embodiments, the processor is further configured to, when the meeting type is the second meeting type, perform grouping processing based on the concept tendency analysis results, the speech initiative score, and the role tendency matrix to determine the discussion group.

[0156] Specifically, in the scenario of a single topic discussion, it is necessary to group users based on the three user attribute analysis results: concept tendency analysis results, speaking enthusiasm score, and role tendency matrix.

[0157] In some embodiments, the complete grouping logic may be as follows: based on a single topic, the idea tendencies of all users (such as support distribution) are counted to determine the discussion goal (collision or detailed discussion). Then, if a collision of ideas is required, participants with different idea tendencies are divided into the same discussion group; if a detailed discussion is required, participants with consistent idea tendencies are divided into the same discussion group. Then, the role ratio and the mean speaking enthusiasm of the members in the discussion group are calculated, and dynamic optimization is performed by exchanging members (such as transferring the redundant highly motivated expressers in a discussion group A to another discussion group B, where the discussion group B is a topic that the highly motivated expressers are more interested in). Finally, the role ratio is checked according to the role tendency matrix, and the missing roles are supplemented (such as deploying hosts from other groups) to ensure that the role structure of each group is complete.

[0158] In this way, if the meeting type is the second type, the video conferencing system performs grouping based on the results of the conceptual analysis, the speaking activity score, and the role orientation matrix to determine the discussion groups. Thus, when discussing a single topic, by integrating multiple user attributes such as conceptual orientation, speaking activity, and role orientation, the grouping ensures a reasonable membership structure within each discussion group, significantly improving the overall efficiency of the meeting process.

[0159] See also Figure 9 In certain embodiments, the method further comprises:

[0160] 017: When the video conference ends, record the user attribute analysis results.

[0161] In some implementations, the determination module is further configured to record the user attribute analysis result when the video conference ends.

[0162] In some embodiments, the processor is further configured to record the user attribute analysis result when the video conference ends.

[0163] Specifically, when a video conference ends, the video conferencing system stores and records the user attribute analysis results generated during the conference using a pre-set user attribute analysis model. This recorded user attribute analysis results can then be used as additional training data to optimize the pre-set user attribute analysis model's feature selection and analysis logic, improving the model's accuracy in identifying user behavior patterns. Furthermore, by retaining historical user attribute analysis results, the video conferencing system can analyze attribute data in subsequent conferences based on long-term user behavior patterns, avoiding the incompleteness of single-conference data and improving the adaptability and accuracy of grouping strategies.

[0164] In this way, when a video conference ends, the video conferencing system records the user attribute analysis results. This post-conference user attribute analysis allows the video conferencing system to perform attribute analysis based on the user's long-term behavior patterns in subsequent conferences, avoiding the one-sidedness of single-conference data and improving the accuracy and adaptability of grouping.

[0165] See also Figure 10 In some embodiments, step 013 (determining user attribute analysis results based on a preset user attribute analysis model and according to participant user performance information and meeting content information) includes:

[0166] 0133: Based on the preset user attribute analysis model, determine the user attribute analysis results according to the recorded historical user attribute analysis results, user performance information of participants and meeting content information.

[0167] In some embodiments, the determination module is further configured to determine the user attribute analysis result based on a preset user attribute analysis model, according to the recorded historical user attribute analysis results, user performance information of the participants, and conference content information.

[0168] In some embodiments, the processor is further configured to determine a user attribute analysis result based on a preset user attribute analysis model, according to recorded historical user attribute analysis results, user performance information of conference participants, and conference content information.

[0169] Specifically, historical user attribute analysis can correct for single-time data biases. For example, if a user's speaking activity score is low in the current meeting due to a device malfunction, but historical data shows that they were highly active in the past 10 meetings, the model will use this historical data to weight the current abnormal score and avoid analysis distortion caused by accidental factors.

[0170] In this way, through the historical user attribute analysis results and the user attribute analysis results determined based on current data, the grouping can take into account the user's long-term behavior patterns and current meeting performance, thereby improving the overall meeting efficiency.

[0171] Based on a pre-set user attribute analysis model, the video conferencing system determines the user attribute analysis results based on historical user attribute analysis results, participant user performance information, and meeting content information. This combination of historical user attribute analysis results avoids analysis bias caused by relying solely on data from a single video conference, ensuring that current user attribute analysis results are more closely aligned with actual user behavior patterns.

[0172] See also Figure 11 In some embodiments, the preset user attribute analysis model is trained by the following steps:

[0173] 021: Send a model training request to the terminal devices of the participants;

[0174] 022: If the model training request is approved, feature extraction is performed on the participant user performance information and meeting content information to determine attribute feature information related to user attributes;

[0175] 023: Send a personal attribute questionnaire to participants at the end of the video conference;

[0176] 024: Determine label data based on the personal attribute survey results obtained based on the personal attribute questionnaire;

[0177] 025: Based on the quantum homomorphic encryption algorithm, the preset user attribute analysis model is trained according to attribute feature information and label data.

[0178] In certain embodiments, the video conferencing grouping device further includes a model training module configured to send a model training request to a participant's terminal device. If the model training request is approved, feature extraction is performed on the participant's user performance information and conference content information to determine attribute feature information related to the user's attributes. Furthermore, upon the conclusion of the video conference, a personal attribute questionnaire is sent to the participant. The video conferencing grouping device further includes a model training module configured to determine label data based on the personal attribute survey results obtained based on the personal attribute questionnaire. Furthermore, based on a quantum homomorphic encryption algorithm, a preset user attribute analysis model is trained based on the attribute feature information and label data.

[0179] In certain embodiments, the processor is further configured to send a model training request to a participant's terminal device. If the model training request is approved, the processor performs feature extraction on the participant's user performance information and meeting content information to determine attribute feature information related to the user's attributes. Furthermore, upon the conclusion of the video conference, the processor sends a personal attribute questionnaire to the participant. The processor is further configured to determine label data based on the personal attribute survey results obtained based on the personal attribute questionnaire. Furthermore, the processor trains a preset user attribute analysis model based on the attribute feature information and label data using a quantum homomorphic encryption algorithm.

[0180] Specifically, a model training request refers to a request sent by the video conferencing system to a participant's terminal device to request training of a preset user attribute analysis model. This request requires authorization from the participant to ensure compliance with subsequent data collection and model training. This prevents unauthorized data use and complies with privacy protection requirements by requesting authorization from the user.

[0181] Feature extraction processing refers to the process of processing the user performance information of participants (such as the first video information, the first voice information and the operation record information) and the conference content information (such as the second video information and the second voice information) on the premise that the model training request is passed, screening out key features related to user attributes (such as interest level and role tendency), and eliminating irrelevant noise data.

[0182] Attribute feature information refers to a set of key features related to user attributes determined through feature extraction processing, such as the frequency of keywords used when the user speaks, the intensity of changes in facial expressions, click preferences in operational behaviors, etc.

[0183] The Personal Attributes Questionnaire is a questionnaire sent to participants after a video conference to collect self-assessment information about their personal attributes. The questionnaire focuses on the user's performance during the conference, such as their speaking enthusiasm, interest in various topics, and their role in the conversation.

[0184] Personal attribute survey results refer to the results obtained based on the personal attribute questionnaires filled out by participants, such as users' ratings of their own speaking enthusiasm, self-assessments of their interest in various topics, and self-assessments of their role preferences.

[0185] Label data refers to the data formed after structured processing of personal attribute survey results, which is used to mark the real attribute category or value of attribute feature information.

[0186] Quantum homomorphic encryption is an encryption technology based on quantum computing theory. It allows calculations to be performed on data in a ciphertext state, and the encrypted calculation results can be obtained without decryption, ensuring the privacy and security of the data during training. In the training of the preset user attribute analysis model, this algorithm is used to encrypt attribute feature information and label data, allowing the training process to be conducted in a ciphertext environment, preventing data leakage or tampering while ensuring the accuracy of the calculation results.

[0187] Training the preset user attribute analysis model refers to optimizing model parameters based on the quantum homomorphic encryption algorithm, using attribute feature information and label data, through machine learning training processes (such as gradient descent, back propagation, etc.), so that the model can accurately output user attribute analysis results based on the input user behavior data and meeting content data.

[0188] In this way, the video conferencing system sends a model training request to the participant's terminal device. If the model training request is approved, the video conferencing system performs feature extraction on the participant's user performance information and meeting content information to determine attribute feature information related to the user's attributes. Then, at the end of the video conference, the video conferencing system sends a personal attribute questionnaire to the participant. The video conferencing system then uses the personal attribute survey results obtained from the personal attribute questionnaire as labeled data. Finally, based on the quantum homomorphic encryption algorithm, the video conferencing system trains the preset user attribute analysis model based on the attribute feature information and labeled data. In this way, through the synergy of user authorization, feature screening, labeled data collection, and quantum encryption technology, the compliance of data collection and user privacy can be guaranteed, and the analytical and reasoning capabilities of the preset user attribute analysis model can be enhanced, thereby improving the accuracy of the output user attribute analysis results.

[0189] See also Figure 12 , an embodiment of the present application provides a video conferencing system 100, the video conferencing system 100 is used to implement the above-mentioned grouping method, the video conferencing system 100 includes a quantum encryption module 110, a user attribute analysis module 120 and a grouping module 130;

[0190] The quantum encryption module 110 can obtain user performance information of participants in the video conference and conference content information of the video conference;

[0191] The user attribute analysis module 120 can determine the user attribute analysis results based on the preset user attribute analysis model and the user performance information of the conference participants and the conference content information;

[0192] The grouping module 130 can group the participants according to the user attribute analysis results when the video conference requires grouping of participants for discussion.

[0193] Specifically, the quantum encryption module 110 obtains the first video information, first voice information, and operation record information of the conference participants through the terminal device and encrypts the data using quantum key distribution technology to ensure transmission security. Furthermore, the second video information and second voice information of the main conference are simultaneously obtained, and the confidentiality of the content data is also guaranteed by quantum encryption technology.

[0194] Next, the user attribute analysis module 120 performs a joint analysis on the decrypted user performance information and the conference content information based on a preset user attribute analysis model to determine a user attribute analysis result.

[0195] Finally, when the video conference requires grouping of participants for discussion, the grouping module 130 groups the participants directly based on the user attribute analysis results output by the user attribute analysis module 120 .

[0196] In this way, the quantum encryption module can obtain user performance information of video conference participants and the meeting content information. Next, the user attribute analysis module can determine the user attribute analysis results based on the user performance information and meeting content information of the participants based on a preset user attribute analysis model. Finally, the grouping module can group participants based on the user attribute analysis results, if the video conference requires group discussion. In this way, by collecting user performance information and meeting content information of participants, combining the user attribute analysis results with the preset user attribute analysis model, and grouping based on the user attribute analysis results, the membership structure within each discussion group is more rationalized, thereby improving the participation and discussion depth of participants in each discussion group.

[0197] The present application also provides a computer-readable storage medium containing a computer program. When the computer program is executed by one or more processors, the one or more processors execute the method of the present application.

[0198] It is understood that a computer program includes computer program code. The computer program code may be in source code form, object code form, executable file, or some intermediate form. Computer-readable storage media may include any entity or device capable of carrying computer program code, recording media, USB flash drives, mobile hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media.

[0199] In the description of this specification, the descriptions with reference to the terms "particularly", "further", "particularly", "understandably", etc. are intended to mean that the specific features, structures, materials or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms are not intended to refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0200] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0201] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A video conference grouping method, characterized in that: The method comprises: Acquire user performance information of participants participating in the video conference, wherein the user performance information of the participants includes first video information, first voice information, and operation record information of the participants; Acquiring conference content information of the video conference, wherein the conference content information includes second video information and second voice information; Determining a user attribute analysis result based on a preset user attribute analysis model and the user performance information of the conference participants and the conference content information; In the case that the video conference requires grouping of the participants for discussion, the participants are grouped according to the user attribute analysis result.

2. The method according to claim 1, characterized in that The method further comprises: Sending an identity authentication request and a rights management request to the terminal device used by the participant; In a case where both the identity authentication request and the authority management request are passed, the user performance information of the conference participant and the conference content information are obtained.

3. The method according to claim 1, characterized in that The obtaining of user performance information of participants in the video conference includes: Receiving encrypted performance information sent by a terminal device used by the participant, wherein the encrypted performance information is obtained by encrypting the user performance information of the participant using a first quantum key obtained by the terminal device based on quantum key distribution technology; Decrypting the encrypted performance information based on the second quantum key obtained by the quantum key distribution technology to determine the user performance information of the participant; The obtaining of the video conference content information includes: Receiving encrypted conference content information sent by the terminal device, wherein the encrypted conference content information is obtained by encrypting the conference content information using a third quantum key obtained by the terminal device based on the quantum key distribution technology; The encrypted conference content information is decrypted based on the fourth quantum key obtained by the quantum key distribution technology to determine the conference content information.

4. The method according to claim 1, wherein The user attribute analysis result includes a first user attribute analysis result and a second user attribute analysis result. The user attribute analysis result is determined based on a preset user attribute analysis model according to the user performance information of the conference participant and the conference content information, including: Based on the preset user attribute analysis model and according to the user performance information of the conference participant, determining the first user attribute analysis result, the first user attribute analysis result including a speaking initiative score and a role tendency matrix; Based on the preset user attribute analysis model, the second user attribute analysis result is determined according to the user performance information of the participant and the conference content information. The second user attribute analysis result includes a topic interest analysis result and a concept tendency analysis result.

5. The method according to claim 4, characterized in that The grouping of the participants according to the user attribute analysis result includes: Based on preset grouping rules, grouping processing is performed according to the meeting type and the user attribute analysis results to determine the discussion group, wherein the meeting type includes a first meeting type and a second meeting type, the first meeting type is used to indicate a meeting that needs to discuss multiple topics, and the second meeting type is used to indicate a meeting that only discusses one topic.

6. The method according to claim 5, characterized in that The method of performing grouping based on the preset grouping rules and the conference type and the user attribute analysis results to determine the discussion group includes: In the case where the conference type is the first conference type, grouping processing is performed based on the topic interest analysis result, the concept tendency analysis result, the speech initiative score and the role tendency matrix to determine the discussion group.

7. The method according to claim 5, characterized in that The method of performing grouping based on the preset grouping rules and the conference type and the user attribute analysis results to determine the discussion group includes: In the case where the meeting type is the second meeting type, grouping processing is performed based on the concept tendency analysis result, the speech initiative score and the role tendency matrix to determine the discussion group.

8. The method according to claim 1, characterized in that The method further comprises: When the video conference ends, the user attribute analysis result is recorded.

9. The method according to claim 7, characterized in that The method of determining a user attribute analysis result based on a preset user attribute analysis model and according to the user performance information of the conference participant and the conference content information includes: Based on the preset user attribute analysis model, the user attribute analysis result is determined according to the recorded historical user attribute analysis results, the user performance information of the conference participants and the conference content information.

10. The method according to claim 1, characterized in that The preset user attribute analysis model is trained by the following steps: Sending a model training request to the terminal device of the participant; If the model training request is approved, performing feature extraction processing on the user performance information of the conference participant and the conference content information to determine attribute feature information related to the user attributes; When the video conference ends, sending a personal attribute questionnaire to the participants; determining label data according to a personal attribute survey result obtained based on the personal attribute questionnaire; Based on the quantum homomorphic encryption algorithm, the preset user attribute analysis model is trained according to the attribute feature information and the label data.

11. A video conferencing system, characterized in that: The video conferencing system is used to implement the grouping method according to any one of claims 1 to 10, and the video conferencing system includes a quantum encryption module, a user attribute analysis module and a grouping module; The quantum encryption module is configured to obtain user performance information of participants in the video conference and conference content information of the video conference; The user attribute analysis module is configured to determine a user attribute analysis result based on a preset user attribute analysis model and according to the user performance information of the conference participants and the conference content information; The grouping module is configured to group the participants according to the user attribute analysis result when the video conference requires grouping the participants for discussion.

Images