Conference management method and device

Abstract

The invention discloses a conference management method and device, and belongs to the technical field of computers. The method comprises the following steps: determining a first issue carried out at a first moment in the process of carrying out an online conference; determining a target user corresponding to the first issue; and prompt information is sent to the electronic equipment of the target user at a second moment to remind the target user to participate in the online conference, and the second moment is later than the first moment.

Description

Technical Field

[0001] The present application belongs to the field of computer technology, and specifically relates to a conference management method and device. Background Art

[0002] As electronic devices become more and more functional, conducting online meetings through electronic devices is becoming more and more common.

[0003] Usually, participants will enter the online meeting when it starts and exit the meeting when it ends. However, some participants may only pay attention to some topics in the meeting, and attending the whole meeting will waste their time. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a conference management method, apparatus, device and medium, which can promptly remind the target user to attend the meeting when the online meeting reaches the first topic corresponding to the target user, without requiring the target user to attend the meeting throughout the whole process, thereby avoiding wasting the target user's time.

[0005] In a first aspect, an embodiment of the present application provides a conference management method, including:

[0006] During the online meeting, determine the first topic to be discussed at the first moment;

[0007] Determine the target users corresponding to the first topic;

[0008] A prompt message is sent to the target user's electronic device at a second time to remind the target user to attend the online meeting, and the second time is later than the first time.

[0009] In a second aspect, an embodiment of the present application provides a conference management device, including:

[0010] A first determination module is used to determine the first topic to be discussed at the first moment during the online meeting;

[0011] A second determination module is used to determine the target users corresponding to the first topic;

[0012] The sending module is used to send a prompt message to the electronic device of the target user at a second time to remind the target user to participate in the online meeting, and the second time is later than the first time.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the first aspect are implemented.

[0014] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0015] In a fifth aspect, an embodiment of the present application provides a chip, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the method described in the first aspect.

[0016] In a sixth aspect, an embodiment of the present application provides a computer program product, which is stored in a storage medium and is executed by at least one processor to implement the method described in the first aspect.

[0017] In the embodiment of the present application, during the online meeting, the first topic to be discussed at the first moment can be determined, and the target user corresponding to the first topic can be determined, and then a prompt message can be sent to the target user's electronic device at the second moment to remind the target user to attend the online meeting. In this way, when the online meeting reaches the first topic corresponding to the target user, the target user can be reminded to attend the meeting in time, without the need for the target user to attend the meeting throughout the entire process, thereby avoiding wasting the target user's time. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a structural block diagram of a conference management system according to an exemplary embodiment;

[0019] Figure 2 This is one of the flowcharts of a conference management method according to an exemplary embodiment;

[0020] Figure 3 This is a second flowchart of a conference management method according to an exemplary embodiment;

[0021] Figure 4 is a structural block diagram of a conference management device according to an exemplary embodiment;

[0022] Figure 5 is a structural block diagram of an electronic device according to an exemplary embodiment;

[0023] Figure 6 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of the present application. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0025] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0026] As mentioned in the background technology, as the functions of electronic devices become more and more abundant, it is becoming more and more common to conduct online meetings through electronic devices.

[0027] Usually, participants will attend the entire meeting, enter the meeting when the online meeting starts, and exit the meeting when the online meeting ends. However, some participants may only focus on some topics in the meeting, especially in large meetings with many topics and many participants. Not every participant needs to participate in all topics. Different topics may require different participants to participate in the discussion. When some participants discuss a certain topic, other participants can only wait, wasting the time of other participants, making the time utilization rate of other participants low. In addition, long-term participation in meetings will waste the energy of participants, resulting in reduced enthusiasm of participants.

[0028] The embodiments of the present application provide a conference management method, apparatus, device, and medium, which can determine the first topic to be discussed at the first moment during an online conference, and determine the target user corresponding to the first topic, and then send a prompt message to the target user's electronic device at the second moment to remind the target user to attend the online conference. In this way, when the online conference reaches the first topic corresponding to the target user, the target user can be reminded to attend the conference in a timely manner without the need for the target user to attend the conference throughout the entire process, thereby avoiding wasting the target user's time and also avoiding the problem of wasting the energy of the participating users due to long-term participation in the conference, which leads to a decrease in the enthusiasm of the participating users.

[0029] The conference management method, apparatus, device and medium provided in the embodiments of the present application are described in detail below with reference to the accompanying drawings through specific embodiments and their application scenarios.

[0030] The conference management method provided in the embodiment of the present application can be applied to the application scenario of conducting online conferences.

[0031] The conference management method provided in the embodiment of the present application can be executed by a conference management system. Figure 1 The conference management system provided in the embodiment of the present application is described in detail.

[0032] Figure 1 The present invention is a structural block diagram of a conference management system according to an exemplary embodiment.

[0033] like Figure 1 As shown, the conference management system 100 may include: a conference server 110 , a judgment center 120 , a conference large model module 130 , a database 140 and a security and privacy protection module 150 .

[0034] The conference large model module 130 may be used to store algorithms required for conference management.

[0035] The database 140 may be used to store data related to current online meetings and historical online meetings.

[0036] The judgment center 120 can be used to determine the first topic to be discussed at the first moment during the online meeting by calling the data in the conference big model module 130 and the database 140; and determine the target users corresponding to the first topic.

[0037] The conference server 110 can be used to send a prompt message to the target user's electronic device at a second time to remind the target user to attend the online meeting. The second time can be later than the first time. The conference server 110 can also be used to obtain the topic of the online meeting, the participating users and other relevant information.

[0038] The security and privacy protection module 150 can be used to protect the data involved in the online meeting.

[0039] Thus, during the online meeting, the first topic to be discussed at the first moment can be determined, and the target user corresponding to the first topic can be determined, and then a prompt message can be sent to the target user's electronic device at the second moment to remind the target user to attend the online meeting. In this way, when the online meeting reaches the first topic corresponding to the target user, the target user can be reminded to attend the meeting in a timely manner, without the need for the target user to attend the meeting throughout the entire process, thereby avoiding wasting the target user's time.

[0040] It should be noted that the above-mentioned execution entities do not constitute a limitation on this application.

[0041] Combine the following Figure 2 The conference management method provided in the embodiment of the present application is described in detail.

[0042] Figure 2 The present invention is a flowchart of a conference management method according to an exemplary embodiment.

[0043] like Figure 2 As shown, the conference management method may include the following steps:

[0044] Step 210, during the online conference, determining a first topic to be discussed at a first moment.

[0045] Here, an online meeting can be a remote meeting conducted through the Internet and communication technology, and users can participate in the online meeting at different locations through computers, mobile phones or other electronic devices. The online meeting can be a voice conference or a video conference.

[0046] The first moment may be the current moment. There may be multiple topics to be discussed in the online meeting. The first topic may be the topic being discussed in the online meeting at the current moment.

[0047] Exemplarily, audio data and / or video data during an online meeting may be obtained, and the audio data and / or video data may be analyzed to determine the topic currently being discussed.

[0048] In some implementations, step 210 may include:

[0049] During the online conference, obtain the voice information of the participating users at the first moment;

[0050] Perform keyword extraction on the text corresponding to the voice information to obtain a first keyword, and perform keyword extraction on multiple topics of the online meeting to obtain second keywords corresponding to the multiple topics respectively;

[0051] Among the second keywords corresponding to the plurality of topics, the topic corresponding to the keyword matching the first keyword is determined as the first topic.

[0052] Here, at the first moment, not all users who should attend the online conference may have joined the conference, and some users may not have entered the conference yet. The users who have joined the conference may be users who have joined the conference at the first moment and are currently attending the online conference.

[0053] The voice information may be the speech content of the users who have participated in the conference. It is understandable that the voice information at a certain moment may be very small, so the voice information at the first moment mentioned here may be the voice information within a period of time, and this period of time includes the first moment. Exemplarily, the voice information may be the speech content of the users who have participated in the conference at the first moment and the preset time before the first moment.

[0054] The multiple topics of the online meeting may be multiple topics that need to be discussed in the online meeting.

[0055] After the voice information is acquired, a topic among the multiple topics that matches the voice information can be determined as the first topic. The first topic matches the voice information can mean that the voice information is the voice information discussing the first topic.

[0056] For example, if multiple topics that need to be discussed in an online meeting include: Topic A "Progress in the application of 5G technology", Topic B "Progress in the development of artificial intelligence mobile phones" and Topic C "Progress in user experience optimization", and the voice message of the participating users at the first moment is "5G technology is currently widely used, and I think it is necessary to upgrade our hardware to support 5G networks", then it can be determined that the first topic matching the voice message is Topic A.

[0057] Since the speeches of the participating users in the online conference are discussions on the topic of the online conference, the speech content can reflect the topic, and the topic being discussed at the first moment can be accurately determined through the voice information of the participating users at the first moment.

[0058] A voice message may include multiple speech contents, so the voice message may correspond to multiple texts, and each text may be processed in the same way.

[0059] The first keyword is a word that can reflect the main content of the voice information. The second keyword is a word that can reflect the main content of the topic.

[0060] Specifically, the voice information may be converted into text first, and then the text may be segmented to obtain a plurality of segmented words, and data cleaning may be performed on the plurality of segmented words according to preset stop words in a preset vocabulary, and segmented words that are the same as the preset stop words may be filtered out, and the remaining segmented words may be unified into a root form by stem extraction or word form restoration to obtain a first keyword. The first keyword may include one or more keywords.

[0061] For example, the voice information may be converted into text through ASR technology.

[0062] For each of the multiple topics, the topic can be segmented to obtain multiple segmented words, and then data cleaning is performed on the multiple segmented words according to the preset stop words in the preset vocabulary, and the segmented words that are the same as the preset stop words are filtered out. The remaining segmented words can also be unified into root forms through stem extraction or word form restoration to obtain the second keyword corresponding to the topic. The second keyword corresponding to a single topic can include one or more keywords.

[0063] Exemplarily, the preset stop words in the preset vocabulary may include words that have no practical meaning, such as "of", "and", "is", "currently", "widely", "I", "feel", "have", "necessary", "next", "we", "with", and "support".

[0064] The first keyword can be matched with the second keyword corresponding to each topic respectively, and the topic corresponding to the second keyword successfully matched with the first keyword is determined as the first topic. Successful matching of the first keyword with the second keyword can mean that the similarity between the first keyword and the second keyword is higher than a fourth threshold, and the fourth threshold can be set according to actual needs.

[0065] For example, the first keywords extracted from the text corresponding to the voice information "Currently 5G technology is widely used, I think it is necessary to upgrade our hardware to support 5G network" may include "5G technology", "upgrade", "hardware" and "5G network". The second keywords extracted from topic A may include "5G technology", "application" and "progress", the second keywords extracted from topic B may include "artificial intelligence", "development" and "progress", and the second keywords extracted from topic C may include "user experience", "optimization" and "progress". After matching the first keyword with the second keywords corresponding to topics A, B and C respectively, it can be determined that the first keyword matches the second keyword corresponding to topic A, so topic A can be determined as the first topic.

[0066] In this way, since the first keyword can reflect the main content of the voice information and the second keyword can reflect the main content of the topic, by matching the first keyword with the second keyword to determine the topic that matches the voice information, the influence of non-main content on the matching accuracy and matching efficiency can be avoided, thereby improving the accuracy and efficiency of determining the topic in progress at the first moment.

[0067] In some implementations, determining, among the second keywords corresponding to the plurality of topics, the topic corresponding to the keyword matching the first keyword as the first topic may include:

[0068] Performing vector conversion on the first keyword to obtain a first feature vector, and performing vector conversion on the second keywords corresponding to the plurality of topics to obtain second feature vectors corresponding to the plurality of topics;

[0069] Among the second feature vectors corresponding to the plurality of topics, the topic corresponding to the feature vector that matches the first feature vector is determined as the first topic.

[0070] Here, the first feature vector may represent the semantic space position of the first keyword, and the second feature vector may represent the semantic space position of the second keyword.

[0071] The first eigenvector and the second eigenvector can both be high-dimensional vectors. The specific dimensions of the eigenvector can be set according to actual needs, and can be set to hundreds or thousands of dimensions if required. In the embodiment of the present application, a 5-dimensional vector is used as an example for introduction.

[0072] Specifically, the first keyword can be converted into a numerical feature vector through word embedding or other text vectorization methods to obtain a first feature vector, and the second keyword can be converted into a numerical feature vector through word embedding or other text vectorization methods to obtain a second feature vector.

[0073] Exemplarily, the keywords may be converted into vectors using a trained word vector model (word2vec) or a bidirectional transformer model (Bidirectional Encoder Representations from Transformers, bert).

[0074] Exemplarily, the first keywords include: “5G technology”, “upgrade”, “hardware” and “5G network”, and the first feature vectors include: the feature vector of “5G technology” [0.12, -0.34, 0.56, -0.78, 0.15], the feature vector of “upgrade” [0.11, -0.15, 0.22, -0.29, 0.08], the feature vector of “hardware” [0.15, -0.21, 0.32, -0.43, 0.10] and the feature vector of “5G network” [0.13, -0.18, 0.28, -0.38, 0.09].

[0075] The second keywords of Topic A include “5G technology”, “application” and “progress”. The second eigenvectors of Topic A include: the eigenvector of “5G technology” [0.12, -0.34, 0.56, -0.78, 0.15], the eigenvector of “application” [-0.25, 0.63, -0.47, 0.29, -0.12] and the eigenvector of “progress” [0.33, -0.24, 0.68, -0.51, 0.21]. The second keywords of Topic B include “artificial intelligence”, “development” and “progress”, and the second feature vector of Topic B includes the feature vector of “artificial intelligence” [0.23, -0.48, 0.36, 0.58, -0.19], the feature vector of “development” [-0.14, 0.52, -0.31, 0.22, 0.41] and the feature vector of “progress” [0.33, -0.24, 0.68, -0.51, 0.21]. The second keywords of Topic C include “user experience”, “optimization” and “progress”, and the second feature vector of Topic C includes the feature vector of “user experience” [0.45, -0.25, 0.67, -0.12, 0.38], the feature vector of “optimization” [0.13, 0.59, -0.44, 0.28, -0.31] and the feature vector of “progress” [0.33, -0.24, 0.68, -0.51, 0.21].

[0076] Then, the first feature vector may be matched with the plurality of second feature vectors respectively, and the topic corresponding to the feature vector in the plurality of second feature vectors that matches the first feature vector may be determined as the first topic.

[0077] In this way, since the feature vector can reflect the semantic spatial position of the keyword, matching the first feature vector and the second feature vector can more accurately determine the semantic distance and similarity between the voice information and the topic, thereby more accurately determining the topic that matches the voice information.

[0078] In some embodiments, the first keyword may include multiple third keywords, the first feature vector may be an average vector of feature vectors corresponding to the multiple third keywords respectively, the second keyword corresponding to each of the multiple topics may include multiple fourth keywords, and the second feature vector corresponding to each of the multiple topics may include third feature vectors corresponding to the multiple fourth keywords respectively;

[0079] Based on this, determining the topic corresponding to the feature vector that matches the first feature vector among the second feature vectors corresponding to the plurality of topics as the first topic may include:

[0080] For each of the multiple topics, perform the following steps:

[0081] respectively determining the similarity between each of the plurality of third feature vectors corresponding to the topic and the first feature vector;

[0082] When the number of target feature vectors is greater than the first threshold, the topic is determined to be the first topic.

[0083] Here, the target feature vector may be a feature vector among a plurality of third feature vectors corresponding to the topic, the similarity of which with the first feature vector is greater than a second threshold.

[0084] The first threshold and the second threshold can be set according to actual needs.

[0085] Exemplarily, the multiple third keywords included in the first keyword are: "5G technology", "upgrade", "hardware" and "5G network", and the feature vectors corresponding to the multiple third keywords are: [0.12, -0.34, 0.56, -0.78, 0.15], [0.11, -0.15, 0.22, -0.29, 0.08], [0.15, -0.21, 0.32, -0.43, 0.10] and [0.13, -0.18, 0.28, -0.38, 0.09]. By calculating the average vector of the feature vectors corresponding to the multiple third keywords, the first feature vector [0.1275, -0.22, 0.345, -0.47, 0.105] can be obtained.

[0086] If the text corresponding to the voice information includes multiple texts corresponding to multiple speeches, then after processing the first keyword corresponding to each text, multiple first feature vectors can be obtained. The multiple first feature vectors can constitute a vector group Ω={ω1,ω2,...,ωm}, where ωi can be the i-th first feature vector (1<=i<=m).

[0087] The second keyword of Topic A includes multiple fourth keywords such as "5G technology", "application" and "progress", the second keyword of Topic B includes multiple fourth keywords such as "artificial intelligence", "development" and "progress", and the second keyword of Topic C includes multiple fourth keywords such as "user experience", "optimization" and "progress". The second feature vector of Topic A includes multiple third feature vectors such as [0.12, -0.34, 0.56, -0.78, 0.15], [-0.25, 0.63, -0.47, 0.29, -0.12] and [0.33, -0.24, 0.68, -0.51, 0.21], and the second feature vector of Topic B includes multiple third feature vectors such as [0.23, -0.48, 0.36, 0.58, -0.19], [- The second eigenvector of topic C includes multiple third eigenvectors: [0.45, -0.25, 0.67, -0.12, 0.38], [0.13, 0.59, -0.44, 0.28, -0.31] and [0.33, -0.24, 0.68, -0.51, 0.21].

[0088] Then, the cosine similarity theorem can be used to calculate the similarity between the vector group Ω and topic A, topic B, and topic C respectively.

[0089] Taking the calculation of the similarity between the vector group Ω and the third eigenvector [0.12, -0.34, 0.56, -0.78, 0.15] of the fourth keyword "5G technology" of topic A as an example, if Ω = [0.066, 0.126, 0.174, -0.276, 0.082], then the similarity between the vector group Ω and the third eigenvector of "5G technology" is cosine_similarity(Ω, "5G technology") = 0.2901 / (0.36525*1.03657)

[0090] ≈0.2901 / 0.3784721175

[0091] ≈0.7667

[0092] Among them, 0.2901 is the dot product of Ω and the third eigenvector of "5G technology", 0.36525 is obtained by taking the modulus operation of Ω, and 1.03657 is obtained by taking the modulus operation of the third eigenvector of "5G technology".

[0093] Based on the above method of calculating similarity, the similarity between the vector group Ω and the third feature vectors corresponding to other fourth keywords of topic A can be technically calculated. For example, it can be recorded as Si(n1,n2,n3.......nx), where nx is the similarity between the vector group Ω and the third feature vector corresponding to the xth fourth keyword of topic A.

[0094] When there are p similarities greater than q in the list Si(n1,n2,n3.......nx), it can be considered that the voice information matches the topic A. P is the first threshold and q is the second threshold.

[0095] In this way, by calculating the similarity between each third feature vector in the second feature vector and the first feature vector, the second feature vector matching the first feature vector can be more accurately determined, thereby accurately determining the topic matching the voice information, that is, the first topic being conducted at the first moment.

[0096] In some implementations, after step 210, the method may further include:

[0097] The first topic is sent to the users who have participated in the online conference and / or the users who are to participate in the conference, so that the electronic devices of the users who have participated in the conference and / or the users who are to participate in the conference display the first topic.

[0098] Here, the users who have participated in the meeting and the users who are to participate in the meeting may be all users who should participate in the online meeting.

[0099] For example, if the first topic is Topic A “Progress in the Application of 5G Technology”, the conference server can send Topic A “Progress in the Application of 5G Technology” to the electronic devices of the participating users and / or the users to be participating, and the electronic devices of the participating users and / or the users to be participating can then display “Current meeting progress: Progress in the Application of 5G Technology”.

[0100] In this way, the meeting progress can be provided to users who have participated and / or users who are about to participate, which increases the transparency of the meeting progress and allows users to clearly understand the meeting progress so that users can prepare in advance before speaking, and it is convenient for users who are about to participate to enter the meeting in time when they are about to speak or discuss topics of interest.

[0101] In some embodiments, the method may further include:

[0102] Convert voice messages during online meetings into text messages;

[0103] Extract meeting keywords from text information;

[0104] Record text messages and meeting keywords.

[0105] Here, the method of extracting conference keywords from text information is the same as the method of extracting the first keyword from the text corresponding to the voice information of the participating user at the first moment, and will not be repeated here.

[0106] In this way, by recording the text information and meeting keywords corresponding to the voice information during the online meeting, the quality of the meeting minutes can be improved, the efficiency of retrieving the meeting content when reviewing the online meeting can be improved, and review and knowledge sharing can be made easier.

[0107] Step 220: Determine the target users corresponding to the first topic.

[0108] Here, the target user may be a user who is interested in the first topic among the users who should participate in the online meeting. The target user may include one or more users.

[0109] Exemplarily, for users who should participate in the online meeting, the user can input topics of interest in advance in the electronic device, and the electronic device sends the topics of interest to the conference server. The conference server can record the topics of interest to each user separately, so that the target users who are interested in the first topic can be determined based on the recorded data.

[0110] In some implementations, step 220 may include:

[0111] Determine the probabilities that the multiple users are interested in the first topic respectively, and obtain the probabilities corresponding to the multiple users respectively;

[0112] A user corresponding to the target probability among multiple users is determined as a target user.

[0113] Here, the multiple users may include at least the users who are to attend the online conference. The multiple users may also include the users who have attended the online conference. Exemplarily, the multiple users may include all the users who should attend the online conference.

[0114] The target probability may be a probability greater than a third threshold among the probabilities corresponding to the plurality of users. The third threshold may be set according to actual needs.

[0115] The probability that each of the multiple users is interested in the first topic can be determined separately. If the probability that a certain user is interested in the first topic is greater than a third threshold, it can be considered that the user is interested in the first topic and needs to be reminded to join the meeting when the first topic is discussed. If the probability that a certain user is interested in the first topic is less than or equal to the third threshold, it can be considered that the user is not interested in the first topic and does not need to be reminded to join the meeting when the first topic is discussed.

[0116] In this way, by analyzing the probability that a user is interested in the first topic, the target users who are interested in the first topic can be determined more accurately.

[0117] In some implementations, respectively determining the probabilities that the plurality of users are interested in the first topic to obtain the probabilities corresponding to the plurality of users may include:

[0118] For each of the multiple users, perform the following steps to obtain the probabilities corresponding to the multiple users:

[0119] The fourth feature vector is input into a behavior model corresponding to the user, and the probability that the user is interested in the first topic is determined based on the fourth feature vector by the behavior model corresponding to the user.

[0120] Here, the fourth feature vector may be a feature vector of keywords extracted from the first topic.

[0121] The behavior model corresponding to the user may be obtained by training a Gaussian mixture model (GMM) based on training samples constructed based on the user's historical behavior data. For example, the user's historical behavior data may include the user's speech information in historical meetings, and the speech information may include language expression, discussion topics, speech length, speech frequency, etc.

[0122] For example, the topics of historical meetings that user Y has attended include "5G technology" and "artificial intelligence", where the feature vector corresponding to "5G technology" is [0.12, -0.34, 0.56, -0.78, 0.15], and the feature vector corresponding to "artificial intelligence" is [0.23, -0.48, 0.36, 0.58, -0.19].

[0123] The feature vector of user Y's speech information on the topic of "5G technology" in the historical meeting is [0.15, -0.30, 0.50, -0.70, 0.20], and the feature vector of the speech information on the topic of "artificial intelligence" is [0.20, -0.45, 0.40, 0.60, -0.25]. The method of determining the feature vector of the speech information is the same as the method of determining the first feature vector corresponding to the voice information in the above embodiment, which will not be repeated here.

[0124] Based on the feature vector [0.15, -0.30, 0.50, -0.70, 0.20] of user Y's speech information on the "5G technology" topic and the "5G technology" topic, a training sample can be constructed. Based on the same method, multiple training samples of user Y can be constructed. Based on multiple training samples of user Y, the GMM model is trained to obtain the behavior model of user Y:

[0125] π1=0.6,μ1=[0.10,-0.35,0.55,-0.75,0.10],Σ1=[[...]],(associated with “5G technology”)

[0126] π2=0.4,μ2=[0.25,-0.40,0.35,0.50,-0.20],Σ2=[[...]],(associated with "artificial intelligence").

[0127] If the first topic is "Internet of Things", and the fourth eigenvector of the "Internet of Things" topic is [0.18, -0.33, 0.47, -0.68, 0.22], then the probability that user Y is interested in the "Internet of Things" topic is:

[0128] P("Internet of Things"|Y)=0.6*N([0.18,-0.33,0.47,-0.68,0.22]|μ1,Σ1)+

[0129] 0.4*N([0.18,-0.33,0.47,-0.68,0.22]|μ2,Σ2)

[0130] Among them, N is the Gaussian distribution function. In order to simplify the calculation, the mean vector μ1:[0.2,-0.3,0.5,-0.7,0.25] is assumed, and the covariance matrix Σ1: is a 5x5 unit matrix, indicating that the features are independent of each other and the variance is 1.

[0131] The multivariate Gaussian distribution formula can be simplified as: N(x|μ,Σ)=(1 / (2π)^(k / 2))*exp(-1 / 2*(x-μ)^T(x-μ))

[0132] The dimension of the feature vector is k = 5, then the Gaussian distribution is:

[0133] N(V|μ1,Σ1)=(1 / (2π)^(5 / 2))*exp(-1 / 2*(V-μ1)*(V-μ1))

[0134] The result of the Gaussian distribution formula can be calculated using simple Python code:

[0135] Fourth eigenvector: vector = np.array([0.18,-0.33,0.47,-0.68,0.22])

[0136] Mean vector: mu1 = np.array([0.2,-0.3,0.5,-0.7,0.25])

[0137] mu2=np.array([0.1,-0.2,0.4,-0.6,0.15])

[0138] Covariance matrix: sigma1 = np.eye(5)

[0139] sigma2 = np.eye(5)

[0140] Calculate the probability density of each distribution at the eigenvector

[0141] p1=multivariate_normal.pdf(vector,mean=mu1,cov=sigma1)

[0142] p2=multivariate_normal.pdf(vector,mean=mu2,cov=sigma2)

[0143] Calculate probability values ​​based on Gaussian mixture model

[0144] interest_probability=0.6*p1+0.4*p2

[0145] print("The probability that user Y is interested in the topic of 'Internet of Things' is:",interest_probability)

[0146] In this way, the probability that the user is interested in the first topic can be accurately calculated through the user's behavior model.

[0147] Step 230, sending a prompt message to the target user's electronic device at the second moment to remind the target user to attend the online meeting.

[0148] Here, the second moment may be later than the first moment, and the duration between the second moment and the first moment may not exceed the preset duration. The duration between the second moment and the first moment may be the duration required for the processing from acquiring voice information to determining the target user.

[0149] The conference server may send a prompt message to the target user's electronic device, which may be a computer, a mobile phone, or other electronic devices.

[0150] The prompt information can be used to remind the target user to attend the online meeting. Specifically, the prompt information can be sent through the application used to conduct the online meeting, or through a phone call, text message, etc.

[0151] In some implementations, after step 220 and before step 230, the method may further include:

[0152] Determine, from the target users corresponding to the first topic, target users to be present but not attending the online meeting at the first moment;

[0153] Based on this, step 230 may include:

[0154] At the second moment, a prompt message is sent to the electronic device of the target user to be a participant in the meeting, so as to remind the target user to be a participant in the meeting to attend the online meeting.

[0155] Here, since the target users may include users who are to attend the meeting and users who have already attended the meeting, and users who have already attended the meeting do not need to be reminded, the target users to attend the meeting who have not attended the online meeting at the first moment can be screened out from the target users, and the target users to attend the meeting are reminded, while non-target users to attend the meeting among the target users will not be reminded.

[0156] The target users to be attended may include users who did not attend the online meeting at the first time. In addition, the target users to be attended may also include users who did not attend the online meeting at a third time, and the third time may be a time between the first time and the second time.

[0157] In this way, users who have not participated in the online meeting at the first moment can be accurately reminded to avoid disturbing users who have participated in the online meeting at the first moment.

[0158] Thus, during the online meeting, the first topic to be discussed at the first moment can be determined, and the target user corresponding to the first topic can be determined, and then a prompt message can be sent to the target user's electronic device at the second moment to remind the target user to attend the online meeting. The second moment is later than the first moment, and the duration between the second moment and the first moment does not exceed the preset duration. In this way, when the online meeting reaches the first topic corresponding to the target user, the target user can be reminded to attend the meeting in time, without the need for the target user to attend the meeting throughout the entire process, thereby avoiding wasting the target user's time.

[0159] In order to better describe the entire solution, based on the above embodiments, a specific example is given. Figure 3 As shown, the conference management method may include steps 301 to 308, which are explained in detail below.

[0160] Step 301, during the online conference, obtain the voice information of the participating users at the first moment.

[0161] Step 302: extract keywords from the text corresponding to the voice information to obtain first keywords, and extract keywords from multiple topics of the online meeting to obtain second keywords corresponding to the multiple topics.

[0162] Step 303 , performing vector conversion on the first keyword to obtain a first feature vector, and performing vector conversion on the second keywords corresponding to the plurality of topics to obtain second feature vectors corresponding to the plurality of topics.

[0163] Step 304: Determine, among the second feature vectors corresponding to the plurality of topics, the topic corresponding to the feature vector that matches the first feature vector as the first topic.

[0164] Step 305 : Determine the probabilities that the multiple users are interested in the first topic respectively, and obtain the probabilities corresponding to the multiple users respectively.

[0165] Step 306: Determine the user corresponding to the target probability among the multiple users as the target user.

[0166] Step 307 , determining target users to be included in the online meeting who have not participated in the online meeting at the first moment from among the target users corresponding to the first topic.

[0167] Step 308: Send a prompt message to the electronic device of the target user to be a participant at the second moment to remind the target user to attend the online meeting.

[0168] The specific process can be found in the above embodiments and will not be described in detail here.

[0169] Thus, during the online meeting, the first topic to be discussed at the first moment can be determined, and the target user corresponding to the first topic can be determined, and then a prompt message can be sent to the target user's electronic device at the second moment to remind the target user to attend the online meeting. The second moment is later than the first moment, and the duration between the second moment and the first moment does not exceed the preset duration. In this way, when the online meeting reaches the first topic corresponding to the target user, the target user can be reminded to attend the meeting in time, without the need for the target user to attend the meeting throughout the entire process, thereby avoiding wasting the target user's time.

[0170] The conference management method provided in the embodiment of the present application can be executed by a conference management device. In the embodiment of the present application, the conference management device provided in the embodiment of the present application is described by taking the conference management method executed by the conference management device as an example.

[0171] Based on the same inventive concept, the present application also provides a conference management device. Figure 4 The conference management device provided in the embodiment of the present application is described in detail.

[0172] Figure 4 The present invention is a structural block diagram of a conference management device according to an exemplary embodiment.

[0173] like Figure 4 As shown, the conference management device 400 may include:

[0174] The first determination module 401 is used to determine the first topic to be discussed at the first moment during the online conference;

[0175] The second determination module 402 is used to determine the target users corresponding to the first topic;

[0176] The sending module 403 is used to send a prompt message to the electronic device of the target user at a second time to remind the target user to participate in the online meeting, and the second time is later than the first time.

[0177] The conference management device 400 is described in detail below, as shown below:

[0178] In one embodiment, the first determining module 401 may include:

[0179] The acquisition submodule is used to obtain the voice information of the participating users at the first moment during the online conference;

[0180] An extraction submodule, configured to extract keywords from the text corresponding to the voice information to obtain a first keyword, and to extract keywords from multiple topics of the online meeting to obtain second keywords corresponding to the multiple topics;

[0181] The first determination submodule is used to determine, among the second keywords corresponding to the plurality of topics, the topic corresponding to the keyword matching the first keyword as the first topic.

[0182] In one embodiment, the first determining submodule may include:

[0183] A conversion unit, configured to perform vector conversion on the first keyword to obtain a first feature vector, and perform vector conversion on the second keywords corresponding to the plurality of topics to obtain second feature vectors corresponding to the plurality of topics;

[0184] The determination unit is used to determine, among the second feature vectors corresponding to the plurality of topics, the topic corresponding to the feature vector that matches the first feature vector as the first topic.

[0185] In one embodiment, the first keyword includes a plurality of third keywords, the first feature vector is an average vector of feature vectors corresponding to the plurality of third keywords respectively, the second keyword corresponding to each of the plurality of topics includes a plurality of fourth keywords, and the second feature vector corresponding to each of the plurality of topics includes the third feature vectors corresponding to the plurality of fourth keywords respectively;

[0186] The determination unit can be specifically used for:

[0187] For each of the multiple topics, perform the following steps:

[0188] respectively determining the similarity between each of the plurality of third feature vectors corresponding to the topic and the first feature vector;

[0189] When the number of target feature vectors is greater than the first threshold, the topic is determined to be the first topic, and the target feature vector is a feature vector whose similarity with the first feature vector is greater than the second threshold among the multiple third feature vectors corresponding to the topic.

[0190] In one embodiment, the second determining module 402 may include:

[0191] The second determination submodule is used to respectively determine the probability that a plurality of users are interested in the first topic, and obtain the probabilities corresponding to the plurality of users, wherein the plurality of users at least include users to be participants of the online conference;

[0192] The third determination submodule is used to determine a user corresponding to a target probability among multiple users as a target user, where the target probability is a probability greater than a third threshold among the probabilities corresponding to the multiple users.

[0193] Thus, during the online meeting, the first topic to be discussed at the first moment can be determined, and the target user corresponding to the first topic can be determined, and then a prompt message can be sent to the target user's electronic device at the second moment to remind the target user to attend the online meeting. In this way, when the online meeting reaches the first topic corresponding to the target user, the target user can be reminded to attend the meeting in a timely manner, without the need for the target user to attend the meeting throughout the entire process, thereby avoiding wasting the target user's time.

[0194] The conference management device in the embodiment of the present application can be an electronic device, or a component in the electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal, or other devices other than a terminal. Exemplarily, the electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted electronic device, a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR) / virtual reality (virtual reality, VR) device, a robot, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc., and can also be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., which is not specifically limited in the embodiment of the present application.

[0195] The conference management device in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0196] The conference management device provided in the embodiment of the present application can achieve Figure 2-3 The various processes implemented in the method embodiment achieve the same technical effect and will not be described again here to avoid repetition.

[0197] In some embodiments, Figure 5 As shown, an embodiment of the present application also provides an electronic device 500, including a processor 501 and a memory 502, wherein the memory 502 stores a program or instruction that can be executed on the processor 501, and when the program or instruction is executed by the processor 501, each step of the above-mentioned conference management method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0198] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0199] Figure 6 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of the present application.

[0200] The electronic device 600 includes but is not limited to: a radio frequency unit 601, a network module 602, an audio output unit 603, an input unit 604, a sensor 605, a display unit 606, a user input unit 607, an interface unit 608, a memory 609, and a processor 610.

[0201] Those skilled in the art will appreciate that the electronic device 600 may also include a power source (such as a battery) for supplying power to each component, and the power source may be logically connected to the processor 610 through a power management system, thereby implementing functions such as managing charging, discharging, and power consumption management through the power management system. Figure 6 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.

[0202] The processor 610 is used to determine a first topic to be discussed at a first moment during the online conference.

[0203] Determine the target users corresponding to the first topic;

[0204] A prompt message is sent to the target user's electronic device at a second time to remind the target user to attend the online meeting, and the second time is later than the first time.

[0205] Thus, during the online meeting, the first topic to be discussed at the first moment can be determined, and the target user corresponding to the first topic can be determined, and then a prompt message can be sent to the target user's electronic device at the second moment to remind the target user to attend the online meeting. In this way, when the online meeting reaches the first topic corresponding to the target user, the target user can be reminded to attend the meeting in a timely manner, without the need for the target user to attend the meeting throughout the entire process, thereby avoiding wasting the target user's time.

[0206] In some embodiments, the processor 610 is further configured to obtain voice information of a user who has participated in the online conference at the first moment during the online conference;

[0207] Perform keyword extraction on the text corresponding to the voice information to obtain a first keyword, and perform keyword extraction on multiple topics of the online meeting to obtain second keywords corresponding to the multiple topics respectively;

[0208] Among the second keywords corresponding to the plurality of topics, the topic corresponding to the keyword matching the first keyword is determined as the first topic.

[0209] In this way, since the first keyword can reflect the main content of the voice information and the second keyword can reflect the main content of the topic, by matching the first keyword with the second keyword to determine the topic that matches the voice information, the influence of non-main content on the matching accuracy and matching efficiency can be avoided, thereby improving the accuracy and efficiency of determining the topic in progress at the first moment.

[0210] In some embodiments, the processor 610 is further configured to perform vector conversion on the first keyword to obtain a first feature vector, and perform vector conversion on the second keywords corresponding to the plurality of topics to obtain second feature vectors corresponding to the plurality of topics;

[0211] Among the second feature vectors corresponding to the plurality of topics, the topic corresponding to the feature vector that matches the first feature vector is determined as the first topic.

[0212] In this way, since the feature vector can reflect the semantic spatial position of the keyword, matching the first feature vector and the second feature vector can more accurately determine the semantic distance and similarity between the voice information and the topic, thereby more accurately determining the topic that matches the voice information.

[0213] In some embodiments, the first keyword includes multiple third keywords, the first feature vector is an average vector of feature vectors corresponding to the multiple third keywords respectively, the second keyword corresponding to each of the multiple topics includes multiple fourth keywords, and the second feature vector corresponding to each of the multiple topics includes the third feature vectors corresponding to the multiple fourth keywords respectively;

[0214] The processor 610 is further configured to perform the following steps for each of the multiple topics:

[0215] respectively determining the similarity between each of the plurality of third feature vectors corresponding to the topic and the first feature vector;

[0216] When the number of target feature vectors is greater than the first threshold, the topic is determined to be the first topic, and the target feature vector is a feature vector whose similarity with the first feature vector is greater than the second threshold among the multiple third feature vectors corresponding to the topic.

[0217] In this way, by calculating the similarity between each third feature vector in the second feature vector and the first feature vector, the second feature vector matching the first feature vector can be more accurately determined, thereby accurately determining the topic matching the voice information, that is, the first topic being conducted at the first moment.

[0218] In some embodiments, the processor 610 is further configured to respectively determine the probability that the plurality of users are interested in the first topic, and obtain the probabilities corresponding to the plurality of users, wherein the plurality of users at least include users to be participants of the online conference;

[0219] A user corresponding to a target probability among multiple users is determined as a target user, where the target probability is a probability greater than a third threshold among the probabilities respectively corresponding to the multiple users.

[0220] In this way, by analyzing the probability that a user is interested in the first topic, the target users who are interested in the first topic can be determined more accurately.

[0221] It should be understood that in the embodiment of the present application, the input unit 604 may include a graphics processor (GPU) 6041 and a microphone 6042, and the graphics processor 6041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 606 may include a display panel 6061, and the display panel 6061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 607 includes a touch panel 6071 and at least one of other input devices 6072. The touch panel 6071 is also called a touch screen. The touch panel 6071 may include two parts: a touch detection device and a touch controller. Other input devices 6072 may include, but are not limited to, a physical keyboard, function keys (such as a volume control button, a switch button, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0222] The memory 609 can be used to store software programs and various data. The memory 609 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory 609 may include a volatile memory or a non-volatile memory, or the memory 609 may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM) and a direct memory bus random access memory (DRRAM). The memory 609 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0223] The processor 610 may include one or more processing units; in some embodiments, the processor 610 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 610.

[0224] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned conference management method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0225] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory, a random access memory, a magnetic disk or an optical disk.

[0226] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned conference management method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0227] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0228] An embodiment of the present application provides a computer program product, which is stored in a storage medium. The program product is executed by at least one processor to implement the various processes of the above-mentioned conference management method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0229] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0230] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0231] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A conference management method, characterized in that: include: During the online meeting, determine the first topic to be discussed at the first moment; Determining target users corresponding to the first topic; Sending a prompt message to the target user's electronic device at a second time to remind the target user to attend the online meeting, the second time being later than the first time.

2. The method according to claim 1, characterized in that During the online meeting, determining the first topic to be discussed at the first moment includes: During the online conference, obtaining voice information of the participating users at the first moment; Perform keyword extraction on the text corresponding to the voice information to obtain a first keyword, and perform keyword extraction on the multiple topics of the online meeting to obtain second keywords corresponding to the multiple topics respectively; Among the second keywords corresponding to the plurality of topics, the topic corresponding to the keyword matching the first keyword is determined as the first topic.

3. The method according to claim 2, characterized in that The step of determining, among the second keywords corresponding to the plurality of topics, a topic corresponding to a keyword matching the first keyword as the first topic, includes: Performing vector conversion on the first keyword to obtain a first feature vector, and performing vector conversion on the second keywords corresponding to the plurality of topics to obtain second feature vectors corresponding to the plurality of topics; Among the second feature vectors corresponding to the plurality of topics, the topic corresponding to the feature vector that matches the first feature vector is determined as the first topic.

4. The method according to claim 3, characterized in that The first keyword includes a plurality of third keywords, the first feature vector is an average vector of feature vectors corresponding to the plurality of third keywords respectively, the second keyword corresponding to each of the plurality of topics includes a plurality of fourth keywords, and the second feature vector corresponding to each of the plurality of topics includes the third feature vectors corresponding to the plurality of fourth keywords respectively; The determining, among the second feature vectors corresponding to the plurality of topics, the topic corresponding to the feature vector that matches the first feature vector as the first topic, includes: For each of the multiple topics, perform the following steps respectively: respectively determining the similarity between each of the plurality of third feature vectors corresponding to the topic and the first feature vector; When the number of target feature vectors is greater than a first threshold, the topic is determined to be the first topic, and the target feature vector is a feature vector among multiple third feature vectors corresponding to the topic whose similarity with the first feature vector is greater than a second threshold.

5. The method according to claim 1, characterized in that The determining the target user corresponding to the first topic includes: Determine respectively the probability that a plurality of users are interested in the first topic, and obtain the probabilities corresponding to the plurality of users, wherein the plurality of users at least include users to be participants of the online conference; A user corresponding to a target probability among the multiple users is determined as the target user, where the target probability is a probability greater than a third threshold among the probabilities respectively corresponding to the multiple users.

6. A conference management device, characterized in that: include: A first determination module is used to determine the first topic to be discussed at the first moment during the online meeting; A second determination module, used to determine target users corresponding to the first topic; The sending module is used to send a prompt message to the electronic device of the target user at a second time to remind the target user to participate in the online meeting, and the second time is later than the first time.

7. The device according to claim 6, characterized in that The first determining module comprises: An acquisition submodule, used for acquiring voice information of the participating users at the first moment during the online conference; An extraction submodule, configured to perform keyword extraction on the text corresponding to the voice information to obtain a first keyword, and to perform keyword extraction on each of the multiple topics of the online meeting to obtain second keywords corresponding to each of the multiple topics; The first determination submodule is configured to determine, among the second keywords corresponding to the plurality of topics, a topic corresponding to a keyword matching the first keyword as the first topic.

8. The device according to claim 7, characterized in that The first determining submodule includes: a conversion unit, configured to perform vector conversion on the first keyword to obtain a first feature vector, and perform vector conversion on the second keywords corresponding to the plurality of topics to obtain second feature vectors corresponding to the plurality of topics; A determination unit is used to determine, among the second feature vectors corresponding to the plurality of topics, the topic corresponding to the feature vector that matches the first feature vector as the first topic.

9. The device according to claim 8, characterized in that The first keyword includes a plurality of third keywords, the first feature vector is an average vector of feature vectors corresponding to the plurality of third keywords respectively, the second keyword corresponding to each of the plurality of topics includes a plurality of fourth keywords, and the second feature vector corresponding to each of the plurality of topics includes the third feature vectors corresponding to the plurality of fourth keywords respectively; The determining unit is specifically used for: For each of the multiple topics, perform the following steps respectively: respectively determining the similarity between each of the plurality of third feature vectors corresponding to the topic and the first feature vector; When the number of target feature vectors is greater than a first threshold, the topic is determined to be the first topic, and the target feature vector is a feature vector among multiple third feature vectors corresponding to the topic whose similarity with the first feature vector is greater than a second threshold.

10. The device according to claim 6, characterized in that The second determining module comprises: A second determination submodule is used to respectively determine the probability that a plurality of users are interested in the first topic, and obtain the probabilities corresponding to the plurality of users, wherein the plurality of users at least include the users to be participants of the online conference; The third determination submodule is used to determine a user corresponding to a target probability among the multiple users as the target user, where the target probability is a probability greater than a third threshold among the probabilities respectively corresponding to the multiple users.

Images