Conference full-process management and control and trusted data tracing system based on AI

Abstract

The invention relates to the technical field of conference management and control, and discloses an AI-based conference full-process management and control and credible data tracing system, which comprises a second conference participating object generation module, a real-time conference theme generation module, a third conference participating object generation module and a credible data generation module. According to the application, the second conference participating object generation module generates a reasonable conference participating object list based on the conference plan and the AI and initiates the conference to ensure that participants are preliminarily matched with the conference theme, and the real-time conference theme generation module updates the conference theme by using the real-time conference multi-modal data, so that the conference discussion is closer to the actual situation, and the conference experience is improved. The third conference participating object generation module optimizes conference participating objects according to a real-time conference theme and improves the conference communication effect, and the credible data generation module analyzes contribution degree recombination data of all the modules and marks the data to achieve credible tracing of the data, so that the conference participating objects and the conference theme are dynamically adjusted according to the conference process, credible management is conducted on the conference data, and the conference communication effect is improved. And the conference efficiency and the data reliability are improved.

Description

Technical Field

[0001] The present application relates to the technical field of meeting control, and specifically to an AI-based full-process meeting control and trustworthy data traceability system. Background Art

[0002] In the context of the increasingly developing meeting technology, as an important activity for information exchange and decision-making, the efficiency of meetings and the importance of data management have become increasingly prominent. However, the inventors found the following problems in the research on existing meeting management technologies:

[0003] The traditional meeting organization method relies on manual screening of participants, often lacking accuracy and making it difficult to ensure a high degree of fit between the participants and the meeting theme, resulting in low meeting discussion efficiency;

[0004] In terms of meeting theme control, there is a lack of a real-time adjustment mechanism, and it is impossible to optimize the theme in a timely manner according to new situations and problems during the meeting process, making the meeting prone to deviating from the core goal;

[0005] At the same time, the meeting data management is chaotic, and multi-modal data (such as video, audio, and text data) has not been effectively integrated and utilized, and it is even more impossible to trace the data trustworthily, making it difficult to guarantee the authenticity and reliability of the data.

[0006] Chinese Patent No. CN119155134A discloses a multi-modal intelligent agent meeting information interaction method, device, equipment, and medium. Although it has innovations in intelligent agent screening and meeting interaction, it focuses on information interaction between multi-modal intelligent agents, the control of the full process of the meeting is not comprehensive enough, lacks a real-time dynamic adjustment mechanism for the meeting theme, and in terms of data management, it only generates meeting reports and does not involve the trustworthy traceability of meeting data, and cannot meet the strict requirements for the authenticity and reliability of meeting data, making it difficult to be applied in scenarios with high requirements for data security and process control.

[0007] In summary, there is an urgent need for a new AI-based technical solution for full-process meeting control and trustworthy data traceability to solve the above technical problems. Summary of the Invention

[0008] The purpose of the present application is to provide an AI-based full-process meeting control and trustworthy data traceability system to solve the technical problems raised in the above background art.

[0009] To achieve the above object, the present application discloses the following technical solutions: A conference full-process control and trustworthy data traceability system based on AI, including a second participant generation module, a real-time conference theme generation module, a third participant generation module, and a trustworthy data generation module that are communicatively connected in sequence. The second participant generation module, the real-time conference theme generation module, and the third participant generation module are communicatively connected;

[0010] The second participant generation module is configured to: obtain a conference plan, use AI to generate an initial conference theme and a first participant based on the conference plan, update the first participant based on the initial conference theme to generate a second participant, and initiate a conference based on the second participant; wherein, the conference plan is formulated by the conference initiator and includes the problems to be solved in the conference and a preliminary list of participants, the initial conference theme is generated from the problems to be solved initially in the conference, the first participant is generated from the preliminary list of participants, and the second participant is a list of participants obtained by optimizing the preliminary list of participants based on the initial conference theme;

[0011] The real-time conference theme generation module is configured to: obtain real-time conference data, and use AI to update the initial conference theme based on the real-time conference data to generate a real-time conference theme; wherein, the real-time conference data is multi-modal data associated with the real-time conference theme generated during the real-time conference process, and this multi-modal data at least includes video data, audio data, and text data, and the real-time conference theme is an optimized conference theme generated by updating the initial conference theme based on the real-time conference process;

[0012] The third participant generation module is configured to: continuously use AI to update the second participant based on the real-time conference theme to generate a third participant, and conduct a conference based on the third participant; wherein, the third participant is a list of participants obtained by optimizing the second participant list based on the real-time conference theme;

[0013] The trustworthy data generation module is configured to: use AI to analyze the contribution of the operation of the second participant generation module, the real-time conference theme generation module, and the third participant generation module to the generation of the real-time conference theme, and use this contribution to reorganize the real-time conference data to generate trustworthy data and mark it, and this mark is used for the traceability of the trustworthy data; wherein, the contribution is used to characterize the influence degree of the change of the participant and / or the change of the real-time conference theme on the real-time conference theme.

[0014] Preferably, the generation of the second participant includes:

[0015] The initial meeting theme is split to obtain corresponding key dimensions. The scores of preliminary participants on these key dimensions are obtained using a preset object dimension table, and the first matching degrees of the preliminary participants and all existing participants with the initial meeting theme are calculated. Based on the ranking of the first matching degrees, the preliminary participants are supplemented to obtain corresponding second participants; wherein, the key dimensions are used to represent different issues discussed during the meeting, the object dimension table stores the scores of the functions of different participants on different key dimensions, and this score is used to represent the professional degree of the participants on the key dimensions. All existing participants are all participants who can attend the meeting. Supplementing the preliminary participants based on the ranking of the first matching degrees means adding the participants with higher rankings to the preliminary participant list.

[0016] Preferably, the generation of the real-time meeting theme includes:

[0017] After obtaining the real-time meeting data, a preset text similarity algorithm is used to calculate the first similarity between the real-time meeting data and the initial meeting theme. The first similarity S 1 is calculated by the formula:

[0018]

[0019] where the number of matching times is the number of words related to the initial meeting theme, and the total number of words is the number of all words obtained from the real-time meeting data. When the first similarity is less than or equal to a preset first similarity threshold, it is determined to trigger the calculation of the second similarity. The second similarity S 2 is calculated by the formula:

[0020]

[0021] where the number of the same keyword is the number of new keyword vocabularies corresponding to the same meeting theme obtained based on the real-time meeting data, and the meeting duration is the acquisition duration corresponding to the real-time meeting data. When the second similarity is greater than or equal to a preset second similarity threshold, the meeting theme corresponding to this keyword vocabulary is determined as the real-time meeting theme.

[0022] Preferably, the generation of the third participants includes:

[0023] The real-time meeting theme is split to obtain the corresponding key dimensions. The scores of the second participants on these key dimensions are obtained using the object dimension table, and the second matching degrees of the second participants and all existing participants with the real-time meeting theme are calculated. Based on the ranking of the second matching degrees, the second participants are supplemented to obtain corresponding third participants.

[0024] Preferably, the generation of the trusted data includes:

[0025] Calculate the corresponding contribution degrees C 1 , C 2 and C 3 of the operations of the second participant generation module, the real-time meeting theme generation module, and the third participant generation module on the generation of the real-time meeting theme, and obtain the parts D 1 , D 2 and D 3 of the real-time meeting data affected by the second participant generation module, the real-time meeting theme generation module, and the third participant generation module. Reorganize the real-time meeting data based on C 1 , C 2 , C 3 , D 1 , D 2 and D 3 to generate the corresponding trusted data.

[0026] Preferably, the calculation of the contribution degree C 1 includes:

[0027] Collect the change values of the matching degrees between each participant and the initial meeting theme before and after the generation of the second participant, and calculate the contribution degree C 1 based on the change values.

[0028] Preferably, the calculation of the contribution degree C 2 includes:

[0029] Calculate the similarity between the real-time meeting theme and the initial meeting theme, and calculate the contribution degree C 2 based on the similarity and the acquisition duration corresponding to the real-time meeting data.

[0030] Preferably, the calculation of the contribution degree C 3 includes:

[0031] Collect the change values of the matching degrees between each participant and the real-time meeting theme before and after the generation of the third participant, and calculate the contribution degree C 3 based on the change values.

[0032] Preferably, the trusted data generation module further includes a contribution degree verification unit;

[0033] The contribution degree verification unit is configured to: verify the rationality of the calculated contribution degree. The verification method is to compare the calculated contribution degree with the contribution degrees of historical similar meetings, and recalculate the contribution degree when the contribution degree deviation is greater than or equal to the preset contribution degree deviation range.

[0034] Preferably, when the trusted data generation module marks the trusted data, the marked content includes the specific value of the contribution degree, as well as the changes corresponding to the real-time meeting theme and the participants.

[0035] Beneficial effects: The AI-based full-process meeting control and trusted data traceability system of the present application realizes the intelligent control of the full process of the meeting and the traceability of trusted data. The second participant generation module generates a reasonable list of participants based on the meeting plan and AI and initiates the meeting to ensure that the participants are initially in line with the meeting theme. The real-time meeting theme generation module updates the meeting theme using real-time meeting multimodal data to make the meeting discussion more in line with the actual situation. The third participant generation module optimizes the participants according to the real-time meeting theme to improve the meeting communication effect. The trusted data generation module analyzes the contribution degrees of each module, reorganizes the data, and marks it to achieve the trusted traceability of the data, so as to dynamically adjust the participants and the meeting theme according to the meeting progress and perform trusted management on the meeting data, solving the problems of inaccurate meeting organization, lagging theme control, and chaotic data management in the prior art, and improving the meeting efficiency and data reliability. Description of the Drawings

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0037] Figure 1 It is a structural block diagram of the AI-based full-process meeting control and trusted data traceability system provided by the embodiment of the present application. Detailed Embodiments

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present application.

[0039] In this article, the term "including" is intended to cover non-exclusive inclusion, so that a process, method, article, or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such a process, method, article, or device. Without more limitations, the elements defined by the statement "including..." do not exclude the existence of additional identical elements in the process, method, article, or device including the said elements.

[0040] This embodiment discloses a conference full - process control and trusted data traceability system based on AI as shown in Figure 1 Figure, which includes a second participant generation module, a real - time conference theme generation module, a third participant generation module, and a trusted data generation module that are sequentially communicatively connected, and the trusted data generation module is communicatively connected to the second participant generation module, the real - time conference theme generation module, and the third participant generation module;

[0041] The second participant generation module is configured to: obtain a meeting plan, use AI to generate an initial conference theme and a first set of participants based on the meeting plan, update the first set of participants based on the initial conference theme to generate a second set of participants, and initiate a meeting based on the second set of participants; among them, the meeting plan is formulated by the meeting initiator and includes the problems to be solved in the meeting and a preliminary list of participants, the initial conference theme is generated from the problems to be solved initially in the meeting, the first set of participants is generated from the preliminary list of participants, and the second set of participants is a list of participants obtained by optimizing the preliminary list of participants based on the initial conference theme;

[0042] The real - time conference theme generation module is configured to: obtain real - time conference data, and use AI to update the initial conference theme based on the real - time conference data to generate a real - time conference theme; among them, the real - time conference data is multi - modal data associated with the real - time conference theme generated during the real - time conference process, and this multi - modal data includes at least video data, audio data, and text data, and the real - time conference theme is an optimized conference theme generated by updating the initial conference theme based on the real - time conference process;

[0043] The third participant generation module is configured to: continuously use AI to update the second set of participants based on the real - time conference theme to generate a third set of participants, and conduct a meeting based on the third set of participants; among them, the third set of participants is a list of participants obtained by optimizing the second set of participants based on the real - time conference theme;

[0044] The trusted data generation module is configured to: use AI to analyze the contribution of the operation of the second participant generation module, the real - time conference theme generation module, and the third participant generation module to the generation of the real - time conference theme, use this contribution to reorganize the real - time conference data to generate trusted data and mark it, and this mark is used for the traceability of the trusted data; among them, the contribution is used to characterize the influence degree of the change of the participants and / or the change of the real - time conference theme on the real - time conference theme.

[0045] Through the above, this embodiment realizes the intelligent control and trustworthy data traceability of the entire meeting process. The second participant generation module generates a reasonable list of participants based on the meeting plan and AI and initiates the meeting to ensure that the participants are initially in line with the meeting theme. The real-time meeting theme generation module updates the meeting theme using real-time meeting multi-modal data to make the meeting discussion more in line with the actual situation. The third participant generation module optimizes the participants according to the real-time meeting theme to improve the meeting communication effect. The trustworthy data generation module analyzes the contribution degrees of each module, reorganizes the data, and marks it to achieve trustworthy traceability of the data. Thus, the participants and the meeting theme are dynamically adjusted according to the meeting progress, and the meeting data is managed trustworthily, solving the problems of inaccurate meeting organization, lagging theme control, and chaotic data management in the prior art, and improving the meeting efficiency and data reliability. It should be noted that this embodiment uses existing AI to realize the intelligent control and trustworthy data traceability of the entire meeting process. Exemplarily, such as natural language processing technology, etc.

[0046] Specifically, the generation of the second participant includes:

[0047] The initial meeting theme is split to obtain the corresponding key dimensions, the scores of the preliminary participants on the key dimensions are obtained using the preset object dimension table, and the first matching degrees of the preliminary participants and all existing participants with the initial meeting theme are calculated. The preliminary participants are supplemented based on the ranking of the first matching degrees to obtain the corresponding second participants; among them, the key dimensions are used to represent different issues discussed during the meeting, and the object dimension table stores the scores of the functions of different participants on different key dimensions, and this score is used to represent the professional degree of the participants on the key dimensions. All existing participants are all the participants who can attend the meeting. Supplementing the preliminary participants based on the ranking of the first matching degrees means adding the participants with higher rankings to the preliminary participant list.

[0048] It should be noted that the key dimensions of this embodiment can be extracted from the meeting theme or determined based on preset rules. Specifically, with the help of AI technology, after in-depth analysis and decomposition of the meeting theme, key dimensions such as for the meeting theme of "increasing the product market share" can be extracted. AI can determine the key dimensions from aspects such as market research, competitor analysis, and product optimization, and use them as an important basis for measuring the matching degree of participants and generating the meeting theme. It can also be corresponding rules formulated based on past meeting experience, industry general standards, or specific business requirements, that is, the key dimensions stored in the object dimension table. Exemplarily, in a business cooperation negotiation meeting, according to past experience, cooperation models, profit distribution, risk assessment, etc. can be set as key dimensions to make the key dimensions match the actual needs of the meeting and ensure the efficient progress of the meeting.

[0049] Through the above, this embodiment realizes more accurate screening and supplementation of the second meeting participants. By calculating the matching degree based on the scores of the preliminary meeting participants in the key dimensions, objects that are more in line with the meeting theme can be found. Moreover, the screening process of the meeting participants is refined, no longer simply relying on simple methods such as label identification, but considering the key dimensions of the meeting theme and the professional level of the objects in depth, so that the selected second meeting participants are more matched with the issues discussed in the meeting in terms of professional ability, further improving the quality of the meeting participants, thus helping to improve the depth and efficiency of the meeting discussion, ensuring that the meeting can better focus on the theme, and promoting the achievement of the meeting goals.

[0050] As a preferred implementation manner of this embodiment, the second meeting participant calculation formula is used to generate the second meeting participants. The second meeting participant calculation formula is:

[0051]

[0052] Where i represents the i-th meeting participant, j represents the j-th key dimension of the initial meeting theme, n represents the number of key dimensions, ω j is the weight of the j-th key dimension, and s i,j is the score of the i-th meeting participant in the j-th key dimension, M i is the first matching degree of the i-th meeting participant calculated. Based on the sorting of M i , the meeting participants are screened and supplemented to achieve the capture of changes in the meeting participants. It can be understood that the technical effects of the third meeting participant generation module in this embodiment are the same as those of the second meeting participant generation module. Therefore, in the third meeting participant generation module, the calculation of the second matching degree will replace ω j and s i,j with the key dimensions of the real-time meeting theme.

[0053] Specifically, the generation of the real-time meeting theme includes:

[0054] After obtaining the real-time meeting data, a preset text similarity algorithm is used to calculate the first similarity between the real-time meeting data and the initial meeting theme. The calculation formula of the first similarity S 1 is:

[0055]

[0056] Where the number of matching times is the number of words related to the initial meeting theme, and the total number of words is the number of all words obtained from the real-time meeting data. When the first similarity is less than or equal to the preset first similarity threshold, it is determined to trigger the calculation of the second similarity. The calculation formula of the second similarity S 2 is:

[0057]

[0058] Among them, the number of the same keywords is the number of new keyword vocabularies corresponding to the same meeting theme obtained based on real-time meeting data, and the meeting duration is the acquisition duration corresponding to the real-time meeting data. When the second similarity is greater than or equal to the preset second similarity threshold, the meeting theme corresponding to the keyword vocabulary is determined as the real-time meeting theme.

[0059] It can be understood that during the meeting, the keyword vocabulary will change with the change of the meeting theme. However, it is found in actual applications that the change of keywords within a short period of time does not represent the change of the meeting theme. Therefore, this embodiment designs the trigger calculation of the second similarity to further optimize the capture of the change of the meeting theme.

[0060] Through the above, this embodiment uses the text similarity algorithm and the similarity calculations in two different stages to achieve the accurate judgment and generation of the real-time meeting theme. When calculating the first similarity, the relevance between the real-time meeting data and the initial meeting theme is initially judged. When it is lower than the threshold, the second similarity calculation is started, and the real-time meeting theme is determined by comprehensively considering factors such as the number of keywords and the meeting duration. This solves the problem of the lack of a real-time theme dynamic adjustment mechanism, can keenly capture new changes during the meeting, timely optimize the meeting theme, keep the meeting always focused on key issues, avoid the discussion deviating from the direction, ensure the timeliness and accuracy of the meeting theme, improve the pertinence and effectiveness of the meeting, and enable the meeting to flexibly adjust the discussion focus according to the actual situation.

[0061] Specifically, the generation of the third participating object includes:

[0062] The real-time meeting theme is split to obtain the corresponding key dimensions, the scores of the second participating object in these key dimensions are obtained by using the object dimension table, and the second matching degree between the second participating object and all existing participating objects and the real-time meeting theme is calculated. Based on the ranking of this second matching degree, the second participating object is supplemented to obtain the corresponding third participating object.

[0063] Through the above, this embodiment realizes the optimized generation of the third participating object. On the basis of the second participating object, the matching degree is obtained by calculating the scores of the second participating object in the key dimensions of the real-time meeting theme, and supplementation is carried out accordingly, making the third participating object more in line with the real-time meeting theme, completing the dynamic adjustment of the participating object based on the key dimensions of the theme, so as to timely adjust the participants according to the theme change during the meeting, ensure that the participants have professional capabilities related to the real-time meeting theme, promote the more efficient progress of the meeting communication, make the meeting discussion more in-depth and valuable, and promote the meeting to better solve practical problems.

[0064] Specifically, the generation of trusted data includes:

[0065] Calculate the corresponding contribution degrees C 1 , C 2 and C 3 of the operation of the second participant generation module, the real-time meeting theme generation module, and the third participant generation module to the generation of the real-time meeting theme respectively, and obtain the parts D 1 , D 2 and D 3 of the real-time meeting data affected by the second participant generation module, the real-time meeting theme generation module, and the third participant generation module. Based on C 1 , C 2 , C 3 , D 1 , D 2 and D 3 reorganize the real-time meeting data to generate the corresponding trusted data.

[0066] Through the above, this embodiment realizes the generation of trusted data. Based on quantifying the contributions of each module to the generation of the real-time meeting theme and then integrating the corresponding data, the generated trusted data is more scientific and reliable. Different from the traditional generation of only meeting reports, this embodiment conducts a more in-depth analysis and processing of the meeting data, can present the meeting data from the perspective of the influence of multiple modules, provides more comprehensive and accurate data support for subsequent meeting evaluation and experience summary, enhances the credibility and usability of the meeting data, and improves the quality of meeting data management.

[0067] As a preferred real-time method of this embodiment, the trusted data calculation formula is used to generate trusted data. The trusted data calculation formula is:

[0068] D t =C 1 *D 1 +C 2 *D 2 +C 3 *D 3

[0069] Among them, D t is the recombined trusted data obtained by calculation. It can be understood that in the generation of traditional meeting reports, the meeting data is recorded in the form of a meeting ledger, but this form will lead to complicated meeting data. Further, the recombination of real-time meeting data based on the contribution degree effectively solves the problem of complicated meeting data.

[0070] Specifically, the calculation of the contribution degree C 1 includes:

[0071] Collect the change values of the matching degrees of each participant object with the initial meeting theme before and after the generation of the second participant object, and calculate the contribution degree C based on these change values 1 。

[0072] Through the above, this embodiment realizes the quantitative evaluation of the contribution of the second participant object generation module. From the perspective of the change in the matching degree between the participant object and the meeting theme, this quantitatively reflects the role of this module in the process of optimizing the participant object list, thereby providing a clear basis for evaluating the effect of the second participant object generation module, helping to further optimize the operation mechanism of this module, improve the accuracy of its screening of participant objects, and indirectly enhance the overall quality and efficiency of the meeting

[0073] As a preferred real-time method of this embodiment, use the first contribution degree calculation formula to generate the contribution degree C 1 The first contribution degree calculation formula is as follows

[0074]

[0075] where m is the number of second participant objects, and ΔM i is the change value of the matching degree of the i-th participant object with the initial meeting theme before and after the generation of the second participant object, and is obtained by using the calculation of the first matching degree mentioned above in specific calculations. It should be noted that the calculation of the contribution degree C in this embodiment 3 adopts the same calculation principle as the first contribution degree calculation formula, and correspondingly collects the number of third participant objects and uses the calculation of the second matching degree mentioned above to obtain the change value of the matching degree with the real-time meeting theme before and after the generation of the third participant object for the calculation of the contribution degree C 3 of the calculation

[0076] Specifically, the calculation of the contribution degree C 2 includes

[0077] Calculate the similarity between the real-time meeting theme and the initial meeting theme, and calculate the contribution degree C based on this similarity and the acquisition duration corresponding to the real-time meeting data 2 。

[0078] Through the above, this embodiment realizes the effective evaluation of the contribution of the real-time meeting theme generation module. This calculation comprehensively considers the degree of theme change and the meeting duration factors, can accurately measure the contribution of this module in the process of updating the meeting theme, thereby quantifying the impact of the real-time meeting theme update on the meeting, providing data support for analyzing the rationality of the meeting theme adjustment, helping to better optimize the real-time meeting theme generation mechanism, making the adjustment of the meeting theme more in line with the actual needs of the meeting, and improving the pertinence and timeliness of the meeting

[0079] As a preferred implementation manner of this embodiment, this embodiment calculates the contribution degree C using the second contribution degree calculation formula 2 , and the second contribution degree calculation formula is as follows:

[0080]

[0081] Among them, S is the similarity between the real-time meeting theme and the initial meeting theme. In this embodiment, the calculation of this similarity can be any existing similarity calculation method, such as calculating the Euclidean distance of the meeting theme, etc. The text will not elaborate here. T is the acquisition duration corresponding to the real-time meeting data and is equal to the meeting duration in the second similarity. It can be understood that in an actual meeting, the content determined as the meeting theme within a short time is more valuable discussion content. Therefore, this embodiment uses the acquisition duration corresponding to the real-time meeting data as a variable to quantify the operation of the real-time meeting theme generation module, thereby achieving the technical effect of improving the quality of the calculation of the contribution degree C 2 of the calculation.

[0082] Specifically, the calculation of the contribution degree C 3 includes:

[0083] Collect the change value of the matching degree between each participant object and the real-time meeting theme before and after the generation of the third participant object, and calculate the contribution degree C based on this change value 3 .

[0084] Through the above, this embodiment realizes the accurate quantification of the contribution of the third participant object generation module. Based on the contribution degree C 1 , further evaluate the contribution of this module from the change in the matching degree between the participant object and the real-time meeting theme, and can intuitively understand its effect on optimizing the participant object, thereby improving the optimization strategy of this module, making the selected third participant object more in line with the requirements of the real-time meeting theme, and further improving the meeting communication effect and meeting quality.

[0085] Specifically, the trusted data generation module further includes a contribution degree verification unit;

[0086] The contribution degree verification unit is configured to: verify the rationality of the calculated contribution degree. The verification method is to compare the calculated contribution degree with the contribution degree of historical similar meetings. When the contribution degree deviation is greater than or equal to the preset contribution degree deviation range, recalculate the contribution degree.

[0087] It should be noted that the contribution degree of the historical similar meetings in this embodiment can be historical data obtained through multiple AI-based full-process meeting management and trusted data traceability, or can also be an empirical value obtained based on the common knowledge of those skilled in the art at the initial stage.

[0088] Through the above, this embodiment realizes the rationality verification of the calculated contribution degree. The contribution degree verification unit compares the calculated contribution degree with the contribution degrees of historical similar meetings. When the deviation is too large, it recalculates, ensuring the accuracy of the contribution degree, effectively avoiding the problem of deviation of reliable data caused by incorrect calculation of the contribution degree, further improving the reliability of reliable data, enhancing the scientificity and rigor of the entire meeting data management system, and providing more reliable data support for meeting decision-making and evaluation.

[0089] Specifically, when the reliable data generation module marks the reliable data, the marked content includes the specific value of the contribution degree and the changes corresponding to the real-time meeting theme and the participating objects.

[0090] Through the above, this embodiment realizes a more comprehensive traceability of reliable data. This marking method provides rich information for subsequent viewing and analysis of reliable data, facilitating the understanding of the background of data generation and the influence of various factors. Different from the traditional relatively simple data marking and traceability, this embodiment enables users to more clearly understand the formation process of meeting data, better mine the data value, provides strong support for summarizing experience and improving processes in meetings, and improves the practicality and effectiveness of meeting data traceability.

[0091] In summary, the AI-based full-process meeting control and reliable data traceability system of this embodiment realizes the intelligent control of the full process of the meeting and the traceability of reliable data. The second participating object generation module generates a reasonable list of participating objects based on the meeting plan and AI and initiates the meeting to ensure that the participants are initially in line with the meeting theme. The real-time meeting theme generation module updates the meeting theme using real-time meeting multi-modal data to make the meeting discussion more in line with the actual situation. The third participating object generation module optimizes the participating objects according to the real-time meeting theme to improve the meeting communication effect. The reliable data generation module analyzes the contribution degrees of each module, reorganizes the data, and marks it to achieve the reliable traceability of the data, thereby dynamically adjusting the participating objects and the meeting theme according to the meeting progress, and performing reliable management of the meeting data, solving the problems of inaccurate meeting organization, lagging theme control, and chaotic data management in the prior art, and improving the meeting efficiency and data reliability.

[0092] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any suitable combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, or other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. The computer-readable storage medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transmission of a computer program from one place to another. The storage media can be any available medium that can be accessed by a computer. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0093] Finally, it should be noted that the above are only the preferred embodiments of the present application and are not used to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An AI-based conference full-process control and trusted data tracing system, characterized by: It includes a second participant object generation module, a real-time conference theme generation module, a third participant object generation module, and a trusted data generation module that are communicatively connected in sequence to the second participant object generation module, the real-time conference theme generation module, and the third participant object generation module; The second participant generation module is configured to: obtain a meeting plan, generate an initial meeting theme and a first participant using AI based on the meeting plan, update the first participant based on the initial meeting theme to generate a second participant, and initiate a meeting based on the second participant; wherein the meeting plan is formulated by the meeting initiator and includes issues to be resolved at the meeting and a preliminary list of participants, the initial meeting theme is generated from the preliminary issues to be resolved at the meeting, the first participant is generated from the preliminary list of participants, and the second participant is a list of participants obtained by optimizing the preliminary list of participants based on the initial meeting theme; The real-time conference theme generation module is configured to: obtain real-time conference data, and use AI to update the initial conference theme based on the real-time conference data to generate a real-time conference theme; wherein the real-time conference data is multimodal data associated with the real-time conference theme generated during the real-time conference, and the multimodal data includes at least video data, audio data, and text data, and the real-time conference theme is an optimized conference theme generated based on the real-time conference process to update the initial conference theme; The third participant generation module is configured to: continuously use AI to update the second participant based on the real-time conference theme to generate a third participant, and conduct a meeting based on the third participant; wherein the third participant is a list of participants obtained by optimizing the second participant list based on the real-time conference theme; The trusted data generation module is configured to: use AI to analyze the contribution of the operation of the second participant generation module, the real-time conference theme generation module and the third participant generation module to the generation of the real-time conference theme, use the contribution to reorganize the real-time conference data to generate trusted data and mark it, and the mark is used for tracing the trusted data; wherein the contribution is used to characterize the degree of influence of changes in participant objects and / or changes in real-time conference themes on the real-time conference theme.

2. The AI-based conference full-process control and trusted data tracing system according to claim 1 is characterized in that: The generation of the second participant object includes: The initial meeting theme is split to obtain corresponding key dimensions, and the scores of the preliminary participants on the key dimensions are obtained using a preset object dimension table, and the first matching degree between the preliminary participants and all existing participants and the initial meeting theme is calculated, and the preliminary participants are supplemented based on the sorting of the first matching degree to obtain corresponding second participants; wherein, the key dimensions are used to characterize different issues discussed during the meeting, and the object dimension table stores scores of different participants' functions on different key dimensions, and the scores are used to characterize the professionalism of the participants on the key dimensions, and the existing all participants are all participants who can attend the meeting, and the sorting based on the first matching degree is used to supplement the preliminary participants to add the participants with higher rankings to the preliminary list of participants.

3. The AI-based conference full-process control and trusted data tracing system according to claim 1 is characterized in that: The generation of the real-time conference topic includes: After acquiring the real-time conference data, a preset text similarity algorithm is used to calculate the first similarity between the real-time conference data and the initial conference topic. The calculation formula of the first similarity S1 is: Among them, the number of matches is the number of words related to the initial meeting topic, and the total number of words is the number of all words obtained from the real-time meeting data. When the first similarity is less than or equal to the preset first similarity threshold, it is determined to trigger the second similarity calculation. The calculation formula of the second similarity S2 is: Among them, the number of identical keywords is the number of new keywords corresponding to the same conference theme obtained based on the real-time conference data, the conference duration is the collection duration corresponding to the real-time conference data, and when the second similarity is greater than or equal to a preset second similarity threshold, the conference theme corresponding to the keyword is determined as the real-time conference theme.

4. The AI-based conference full-process control and trusted data tracing system according to claim 2 is characterized in that: The generation of the third participant object includes: The real-time conference topic is split to obtain the corresponding key dimension, the score of the second participant on the key dimension is obtained by using the object dimension table, and the second matching degree of the second participant and all existing participants with the real-time conference topic is calculated, and the second participant is supplemented based on the sorting of the second matching degree to obtain the corresponding third participant.

5. The AI-based conference full-process control and trusted data tracing system according to claim 1 is characterized in that: The generation of the trusted data includes: The corresponding contribution degrees C1, C2 and C3 of the operation of the second participant generation module, the real-time conference theme generation module and the third participant generation module to the generation of the real-time conference theme are calculated respectively, and the real-time conference data parts D1, D2 and D3 under the influence of the second participant generation module, the real-time conference theme generation module and the third participant generation module are obtained, and the real-time conference data are reorganized based on C1, C2, C3, D1, D2 and D3 to generate corresponding trusted data.

6. The AI-based conference full-process control and trusted data tracing system according to claim 5 is characterized in that: The calculation of contribution C1 includes: The change value of the matching degree between each participant and the initial conference theme before and after the second participant is generated is collected, and the contribution degree C1 is calculated based on the change value.

7. The AI-based conference full-process control and trusted data tracing system according to claim 5 is characterized in that: The calculation of contribution C2 includes: The similarity between the real-time conference topic and the initial conference topic is calculated, and the contribution C2 is calculated based on the similarity and the collection duration corresponding to the real-time conference data.

8. The AI-based conference full-process control and trusted data tracing system according to claim 5 is characterized in that: The calculation of contribution C3 includes: The change value of the matching degree between each participant and the real-time conference theme before and after the third participant is generated is collected, and the contribution degree C3 is calculated based on the change value.

9. The AI-based conference full-process control and trusted data tracing system according to claim 1 is characterized in that: The trusted data generation module also includes a contribution verification unit; The contribution verification unit is configured to verify the rationality of the calculated contribution by comparing the calculated contribution with the contribution of similar historical meetings, and recalculating the contribution when the contribution deviation is greater than or equal to a preset contribution deviation range.

10. The AI-based conference full-process control and trusted data tracing system according to claim 1 is characterized in that: When the trusted data generation module marks the trusted data, the marking content includes the specific value of the contribution and the corresponding changes of the real-time conference theme and the participants.

Images