Instant messaging group management optimization method and system

Abstract

The present invention relates to the field of instant messaging monitoring technology, and specifically discloses an instant messaging group management optimization method and system, comprising: obtaining a set of interaction data of group members, obtaining interaction information of multiple members based on the interaction data set, wherein the member interaction information includes message content information and attachment type information; obtaining the response time interval with other members based on the message content information, and obtaining the response activity based on the response time interval; obtaining the type information correlation with other members based on the attachment type information; and obtaining a real-time authority index based on the response activity and type information correlation. In terms of data processing, the present invention obtains comprehensive interaction data of group members, solves the problem of incomplete collection and chaotic organization of traditional management data, and enables managers to grasp the group dynamics, such as understanding project discussions and information sharing in enterprise project groups.

Description

Technical Field

[0001] The present invention relates to the field of instant messaging technology, and in particular to an instant messaging group management optimization method and system. Background Art

[0002] With the rapid development of Internet technology, instant messaging groups have been widely used in various scenarios. Members of different departments use instant messaging groups to communicate projects, assign tasks and report progress.

[0003] Existing management methods only capture basic information such as the number of messages and the number of times members have spoken, making it difficult to deeply analyze message content. For example, in an enterprise project discussion group, managers are unable to quickly and accurately extract key points regarding key technical issues from a flood of messages, nor can they systematically organize shared attachments such as technical documents, leading to erroneous control over project progress. This makes it difficult for managers to comprehensively grasp the group's communication status and information flow, hindering their ability to effectively utilize information to drive progress. Summary of the Invention

[0004] The purpose of the present invention is to provide an instant messaging group management optimization method and system to solve the technical problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for optimizing instant messaging group management, comprising:

[0007] Acquire a group member interaction data set, and acquire a plurality of member interaction information according to the interaction data set, wherein the member interaction information includes message content information and attachment type information;

[0008] Obtaining a response time interval with other members based on the message content information, and obtaining a response activity based on the response time interval;

[0009] Obtaining a degree of association with other members' type information based on the attachment type information;

[0010] Obtaining a real-time permission index based on the response activity and type information correlation;

[0011] Obtaining a message authority level according to a real-time authority index corresponding to the member interaction information, and obtaining a message prediction value according to the message authority level;

[0012] A resource allocation value is obtained according to the message prediction value, and member interaction information is regulated based on the resource allocation value.

[0013] Preferably, the step of obtaining the response time interval with other members according to the message content information, and obtaining the response activity according to the response time interval includes:

[0014] Acquire multiple conversation threads according to the message content information;

[0015] Obtaining a timestamp according to the conversation thread;

[0016] Generate a time series sorted list of each message in the conversation thread according to the timestamp mark;

[0017] Obtaining time interval differences between adjacent messages according to the time series sorted list, and generating a response time interval set;

[0018] A response activity value is obtained according to the response time interval set.

[0019] Preferably, the step of obtaining the degree of association with type information of other members based on the attachment type information includes:

[0020] Acquire multiple group classification labels according to the attachment type information;

[0021] Establishing a group topic-attachment type mapping table according to the plurality of group classification tags;

[0022] Obtaining the frequency of information transmission of similar attachments within the current session period according to the group topic-attachment type mapping table;

[0023] The type information association is obtained according to the information sending frequency.

[0024] Preferably, the step of obtaining a real-time permission index according to the response activity and the correlation between the type information includes:

[0025] Obtaining message keywords according to the message content information;

[0026] Get the frequency probability of the message keywords;

[0027] Obtaining message activity based on the word frequency probability and response activity;

[0028] A real-time authority index is obtained according to the message activity and type information correlation.

[0029] Preferably, the step of obtaining a message authority level according to the real-time authority index corresponding to the member interaction information, and obtaining a message prediction value according to the message authority level includes:

[0030] Acquire the corresponding number of participating members according to the member interaction information;

[0031] Get the total number of group members;

[0032] Obtaining the message permission level based on the number of participating members, the total number of participating members, and the real-time permission index;

[0033] Acquiring content density of information according to the member interaction information;

[0034] Obtaining a predicted message reading value based on the content density of the information;

[0035] A message prediction value is obtained according to the predicted message reading value.

[0036] Preferably, the step of obtaining a resource allocation value according to the message prediction value includes:

[0037] Obtaining a message differentiated transmission control value and a resource requirement value according to the message prediction value;

[0038] Obtaining a first weight coefficient according to the message differentiated transmission control value;

[0039] Obtaining a second weight coefficient according to the resource demand value;

[0040] Acquire a resource allocation value according to the message differentiated transmission control value, the first allocation weight coefficient, the resource requirement value, and the second allocation weight coefficient;

[0041] The present invention also provides an instant messaging group management optimization system, comprising:

[0042] a first acquisition module, configured to acquire an interaction data set of group members, and acquire a plurality of member interaction information according to the interaction data set, wherein the member interaction information includes message content information and attachment type information;

[0043] A second acquisition module is used to acquire a response time interval with other members according to the message content information, and acquire a response activity according to the response time interval;

[0044] A third acquisition module is used to acquire the degree of association of type information with other members based on the attachment type information;

[0045] A fourth acquisition module is configured to acquire a real-time permission index based on the response activity and the correlation between the type information;

[0046] a fifth acquisition module, configured to acquire a message authority level according to a real-time authority index corresponding to the member interaction information, and acquire a message prediction value according to the message authority level;

[0047] The control module is used to obtain a resource allocation value according to the message prediction value, and to control member interaction information based on the resource allocation value.

[0048] Preferably, the second acquisition module includes:

[0049] A first acquiring unit, configured to acquire multiple conversation threads according to the message content information;

[0050] A second acquiring unit, configured to acquire a timestamp according to the conversation thread;

[0051] A third acquiring unit is configured to generate a time series sorted list of each message in the conversation thread according to the timestamp mark;

[0052] a fourth acquiring unit, configured to acquire time interval differences between adjacent messages according to the time series sorted list, and generate a response time interval set;

[0053] A fifth acquiring unit is configured to acquire a response activity value according to the response time interval set.

[0054] Preferably, the third acquisition module includes:

[0055] a sixth acquiring unit, configured to acquire a plurality of group classification labels according to the attachment type information;

[0056] a seventh acquiring unit, configured to establish a group subject-attachment type mapping table according to the plurality of group classification tags;

[0057] An eighth acquiring unit, configured to acquire, according to the group topic-attachment type mapping table, the information sending frequency of the same type of attachments in the current session period;

[0058] The ninth acquiring unit is configured to acquire the type information association degree according to the information sending frequency.

[0059] Preferably, the fourth acquisition module includes:

[0060] a tenth acquiring unit, configured to acquire message keywords according to the message content information;

[0061] An eleventh obtaining unit, configured to obtain a word frequency occurrence probability of a message keyword;

[0062] a twelfth acquiring unit, configured to acquire message activity according to the word frequency occurrence probability and the response activity;

[0063] The thirteenth obtaining unit is used to obtain a real-time authority index according to the message activity and the correlation between the type information.

[0064] The beneficial effects of the present application are as follows: in terms of data processing, the present invention obtains comprehensive interaction data of group members, solves the problems of incomplete collection and chaotic organization of traditional management data, and enables managers to grasp the dynamics of the group, such as understanding project discussions and information sharing in an enterprise project group. In terms of member activity assessment, the response activity is obtained through the response time interval, which can distinguish differences in member participation, such as in a group of game enthusiasts, the enthusiasm of members can be clearly identified. For attachment management, a mapping table is established to obtain the correlation of type information, which solves the problems of chaotic attachment management and untargeted information dissemination. The design team group can optimize resource classification accordingly. In terms of authority allocation, a real-time authority index is derived by combining multiple factors to make authority allocation more fair and reasonable, and promote the development of open source projects. At the message management level, message authority levels and predicted values are obtained to avoid important messages being overwhelmed. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a schematic diagram of a method flow chart according to an embodiment of the present application.

[0066] Figure 2 This is a schematic diagram of the system structure of an embodiment of the present application.

[0067] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0068] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0069] like Figure 1 As shown, the present application provides an instant messaging group management optimization method, comprising:

[0070] S1. Acquire a group member interaction data set, and acquire a plurality of member interaction information based on the interaction data set, wherein the member interaction information includes message content information and attachment type information;

[0071] S2. Obtaining a response time interval with other members based on the message content information, and obtaining a response activity based on the response time interval;

[0072] S3. Obtaining a degree of correlation with type information of other members based on the attachment type information;

[0073] S4. Obtaining a real-time permission index based on the response activity and the correlation between the type information;

[0074] S5. Obtaining a message authority level according to the real-time authority index corresponding to the member interaction information, and obtaining a message prediction value according to the message authority level;

[0075] S6. Obtain a resource allocation value according to the message prediction value, and regulate member interaction information based on the resource allocation value.

[0076] As described in steps S1-S6 above, existing technologies are often unable to fully collect the interaction data of group members. They can only obtain some basic information, such as the number of messages, the number of times members speak, etc., and lack in-depth analysis of message content and systematic organization of attachment types. This makes it difficult for managers to grasp the communication situation and information flow of the group as a whole. For example, in an enterprise project discussion group, managers cannot accurately understand the discussion details of key technical issues of the project, as well as the frequency and effect of sharing related technical documents, resulting in deviations in the control of project progress;

[0077] The present invention captures a collection of group member interaction data and extracts message content and attachment type information from it, providing a solid data foundation for subsequent management. This step not only addresses the incomplete and disorganized data collection issues inherent in traditional management, but also enables managers to fully understand group dynamics. For example, in an enterprise project discussion group, this data allows managers to clearly understand members' discussions on project progress and technical challenges, as well as the types of shared materials, thereby better organizing and coordinating work. Based on the data captured in step S1, step S2 calculates the response time intervals with other members based on message content information, thereby calculating response activity. This step effectively measures member participation and enthusiasm in group communication, overcoming the drawback of traditional management that makes it difficult to distinguish differences in member participation. Through a series of operations, including identifying conversation threads, adding timestamps, generating a time series sorted list, and calculating time interval differences, a precise response activity value is derived. For example, in a game enthusiast group, this process can clearly identify differences in member participation enthusiasm, allowing administrators to focus on suggestions from active members while encouraging less active members to participate. Step S3 then uses attachment type information to determine the correlation between type information and other members. This approach addresses the chaotic attachment management and lack of targeted information dissemination in traditional group management. By adding categorized tags, establishing a mapping table between group topics and attachment types, and counting the frequency of similar attachments sent, it uncovers group members' preferences and connections regarding information sharing. In design team groups, administrators can optimize resource classification and recommendations based on this correlation, creating dedicated folders for frequently shared attachment types and pushing relevant information to improve information dissemination efficiency. Next, step S4 combines the response activity and type information correlation obtained in steps S2 and S3 to generate a real-time permission index. This index comprehensively considers members' actual contributions and performance, breaking away from the traditional model of assigning permissions based solely on membership status and making permission allocation more equitable and reasonable. In open source project development groups, members who actively participate in discussions and frequently share valuable code files can, with a higher real-time permission index, be granted write access to the code repository, effectively promoting efficient project development. Based on the real-time permission index, step S5 further determines message permission levels and message prediction values. This step addresses the difficulty in distinguishing message importance and the tendency for important messages to be buried in traditional management. By counting the number of participating members and the total number of group members, and combining this with a real-time permission index to calculate message permission levels, and analyzing information content density to assess message reading value, a message prediction value is derived. In financial investment groups, messages with high permission levels are prioritized. The message prediction value helps administrators plan ahead and ensure that important messages are addressed promptly. Finally, step S6 uses the message prediction value to determine resource allocation values, which are then used to regulate member interaction information. This effectively addresses the problem of traditional resource allocation being lacking in specificity and prone to waste or insufficient resources.By determining the message's differentiated transmission control value, resource requirement, and corresponding weight coefficients, resource allocation can be accurately calculated, enabling precise resource allocation. For example, within a software development project group, technical support resources and server bandwidth can be rationally allocated for critical messages related to key feature development, ensuring smooth project progress.

[0078] In one embodiment, the step of obtaining a response time interval with other members based on the message content information, and obtaining a response activity based on the response time interval, includes:

[0079] S201, obtaining multiple conversation threads according to the message content information;

[0080] S202: Obtain a timestamp according to the conversation thread;

[0081] S203: Generate a time series sorted list for each message in the conversation thread according to the timestamp mark;

[0082] S204. Obtain time interval differences between adjacent messages according to the time series sorted list, and generate a response time interval set;

[0083] S205: Obtain a response activity value according to the response time interval set.

[0084] As described in steps S201-S205 above, the present invention obtains multiple conversation threads based on message content information. In traditional instant messaging groups, messages are disorganized, with discussions on different topics intertwined, making it difficult for members to quickly locate the information they need. However, through this step, using natural language processing technology and keyword matching algorithms, based on a pre-defined keyword library and combined with context, messages within the group can be rationally categorized and organized into different conversation threads. Just like in an e-commerce operations group, messages regarding product promotions, customer feedback, logistics issues, and other issues can be categorized separately. This creates a clearer and more organized message structure, greatly improving the efficiency of information acquisition and communication. Next, step S202 obtains timestamps based on the conversation threads. Because traditional management methods lack detailed message time recording, in-depth temporal analysis is difficult. In this step, once a message is accurately categorized into the corresponding conversation thread, the system automatically obtains the precise time the message was sent and records it as a timestamp in international standard time format, accurate to the second or even millisecond. For example, in an education and teaching discussion group, adding timestamps to discussion topics posted by teachers and subsequent member replies not only clearly demonstrates the discussion process, but also allows for checking the execution of time-limited replies, providing an accurate basis for time series analysis. Then, step S203 generates a time series sorted list for each message in the conversation thread based on the timestamp. In the past, the order in which messages were displayed was often disrupted by factors such as network latency and device refreshes, resulting in a chaotic time sequence and inability to accurately analyze the discussion process. In this step, the system uses algorithms such as quick sort or merge sort to rearrange messages in the same conversation thread based on timestamps. Taking a technology product discussion group as an example, the sorted time series sorted list generated can intuitively show the changes in discussion popularity at different times after the product release, making it easier for administrators to grasp discussion trends and provide strong support for adjusting management strategies. In the subsequent step S204, the time interval difference between adjacent messages is obtained based on the time series sorted list, and a response time interval set is generated. Traditional management makes it difficult to accurately measure the response speed of members and cannot accurately judge their contributions to the discussion. In this step, the system traverses the time series sorted list, calculates the time difference between adjacent messages in turn, collects all the differences to form a response time interval set, and cleans and repairs abnormal data. For example, in a medical emergency knowledge exchange group, this method can quantify the differences in members' response speed to urgent issues, providing objective data indicators for evaluating members' participation. Finally, step S205 obtains the response activity value based on the response time interval set, obtains the number of valid replies based on the response time interval set, obtains the delay time based on the number of valid replies, and obtains the time weight coefficient based on the delay time. The response activity value calculation formula is:

[0085] ;

[0086] in, Indicates the response activity value, Indicates the Number of valid responses, Indicates the Delay time, Indicates the time weight coefficient; in the optimization of instant messaging group management, the response activity value calculation formula is " " is used to measure the members' enthusiasm for participating in group communication. In the formula, Represents the nth number of valid replies, reflecting the member's actual contribution to the discussion; is the nth delay time, that is, the interval from the message sending to the receipt of the nth valid reply, reflecting the timeliness of the reply; S is the time weight coefficient, which is determined according to the delay time. The shorter the delay time, the larger S is, and the greater the impact on the response activity value. When calculating, first calculate each valid reply , As the denominator, it is reasonable to reduce the contribution of replies with long delays to the activity value. The sum of the values of all valid replies and multiplication by the time weight coefficient S can comprehensively reflect the participation and activity of members in the entire discussion process. For example, in an online education exchange group, the teacher posted a discussion on a difficult problem. Student 1 replied with 4 valid solutions within 5 minutes. At this time, =4, = 5 minutes. Assuming that according to the set rules, S=0.8 for 5 minutes, then the contribution of this part to the response activity value is Student 2 responded with 2 valid responses within 10 minutes. =2, = 10 minutes, corresponding to S = 0.6, and its contribution is Student 3 responded once in 15 minutes. =1, = 15 minutes, S = 0.4, the contribution is . Adding up these contribution values, the response activity value of this discussion is X=0.53+0.11+0.025=0.665. Using the response activity value calculated by this formula, teachers can intuitively understand students' participation in problem discussions, praise students with high activity, and encourage students with low activity to participate, thereby optimizing the learning and communication atmosphere of the group and improving learning outcomes. This calculation method can comprehensively and scientifically evaluate the response activity of members in the group. Managers can understand member participation based on the activity value, reasonably assign tasks, give rewards, supervise or train inactive members, and improve group management efficiency.

[0087] Traditional management lacks a scientific, quantitative assessment of member engagement, making it difficult to fairly evaluate member contributions. At this step, a specific algorithm, such as a weighted average of the inverse of the response interval, is used to determine a response activity value. For example, in a sales team group, this value can quickly identify members who actively and passively participate in discussions. Managers can then use this information to reward active members and provide training and support to passive members, thereby optimizing group management decisions.

[0088] In one embodiment, the step of obtaining the degree of association with type information of other members based on the attachment type information includes:

[0089] S301, obtaining multiple group classification labels according to the attachment type information;

[0090] S302: Create a group topic-attachment type mapping table based on the plurality of group classification tags;

[0091] S303: Obtain the frequency of information transmission of similar attachments in the current session period according to the group topic-attachment type mapping table;

[0092] S304: Acquire type information relevance according to the information sending frequency.

[0093] As described in steps S301-S304 above, the present invention obtains multiple group classification tags based on attachment type information. In traditional instant messaging group management, attachments are often shared in a disorderly manner and lack effective classification methods. This makes it difficult for members to find specific attachments and greatly reduces the efficiency of information dissemination. Step S301, however, leverages pre-set classification rules and keyword recognition technology to establish a keyword database for common attachment types, automatically extracting relevant information about the attachment and matching it with the database. For special attachments that are difficult to match, the administrator manually adds tags, thereby achieving a preliminary classification and summary of the attachments. For example, in a design company project group, by obtaining classification tags such as "design materials," "project documents," and "progress reports," members can quickly locate information when searching, greatly improving the targeted nature of information retrieval. Next, step S302 establishes a group topic-attachment type mapping table based on the multiple group classification tags. Under traditional management models, the relationship between group topics and attachment types is unclear, resulting in a lack of clear direction when members share attachments. This leads to a large number of irrelevant attachments flooding the group, disrupting normal discussion and collaboration. In step S302, the system uses natural language processing technology to extract topic keywords from historical group discussion records and, based on attachment sharing, establishes a correlation between the two. For example, in a research project group, around the topic of "experimental data analysis," a mapping table can clearly display commonly used attachment types such as "experimental data table (CSV)" and "experimental analysis report (PDF)." This allows members to more specifically prepare and find information when participating in discussions, strengthening information sharing and collaboration around the topic. Then, in step S303, the frequency of information transmission of similar attachments within the current session cycle is obtained based on the group topic-attachment type mapping table. Traditional management methods cannot track the dynamic changes in attachment sharing within the group in real time, making it difficult to determine which information is truly important and urgently needed within a specific time period. Step S303, however, uses the preset session cycle as the time unit to monitor attachment upload operations in real time and compile statistics on attachment sharing records within the current session cycle based on the mapping table. For example, in a marketing activity planning group, by counting the frequency of sending "promotional posters (JPEG)" during the planning period of a new product promotion activity, if the sending frequency is high, it can be determined that poster design and modification are the current work priorities. Based on this, members can pay attention to relevant discussions and attachments in a timely manner, providing strong support for project advancement. Finally, step S304 obtains the type information relevance based on the information sending frequency, obtains the preset relevance value, and obtains the preset sending frequency. The calculation formula is:

[0094] ;

[0095] in, Indicates the type information association, Indicates the frequency of information sending. Indicates the preset sending frequency. Indicates a preset association value; in step S304 of obtaining the type information association degree, " This calculation formula plays a key role, which quantifies the relevance of the attachment to the group topic and members by taking into account multiple factors. Represents the correlation between type information. Its value directly reflects the degree of correlation between the attachment type and the group topic. The higher the value, the closer the correlation. The frequency of information sending records the actual number of times the same type of attachment is sent during the current session, reflecting the level of activity of this type of attachment in group discussions. The preset sending frequency is a reference value set based on the group's historical data, business needs, etc., and is used to measure the actual sending frequency. It is a preset correlation value, which reflects the basic correlation between the attachment type and the group topic under ideal conditions, and its value ranges from 0 to 1. This function adjusts the impact of message sending frequency on relevance. When P is greater than M, the actual sending frequency is higher than expected. While this indicates that the attachment is receiving attention, such a high frequency may indicate an anomaly, and the magnitude of the relevance increase will be appropriately reduced. When P is less than M, the relevance decreases, but the magnitude of the decrease is adjusted based on the difference. When P is equal to M, the relevance is primarily determined by the preset relevance value, I. For example, in an e-commerce operations group promoting new clothing, if the preset sending frequency, M, for "product image (JPEG)" is 80 times, the preset relevance value, I, is 0.9, and the actual sending frequency, P, is 100 times, then O = (1 - 0.1 × (100 - 80)) × 0.9 = 0.9 (taking the absolute value). This indicates that "product image (JPEG)" is highly relevant to the group theme in the current new clothing promotion campaign. Based on this information, administrators can rationally allocate resources, such as providing more storage space for high-definition images and increasing image transmission bandwidth. Furthermore, in terms of information recommendations, messages containing product images are prioritized, effectively promoting the new clothing promotion. This fully demonstrates the importance of this formula in practical group management. The traditional management model cannot achieve accurate resource allocation and information recommendation due to the lack of quantitative assessment of the degree of correlation between attachments and group topics and members. In S304, the system uses a specific algorithm to normalize the frequency of information sending, and performs a comprehensive calculation based on relevant factors such as whether the attachments are downloaded and viewed. For example, in an e-commerce operation group, the frequency of sending and the number of views of "product images (JPEG)" and "product introduction documents (PDF)" are standardized and weighted to obtain the type information correlation value between them and the current group topic. Based on these values, managers can judge that product images are more critical to business advancement at the current stage, and then make more reasonable decisions on resource allocation and information recommendation. Members can also filter and focus on valuable information more efficiently.

[0096] In one embodiment, the step of obtaining a real-time permission index based on the response activity and the correlation between the type information includes:

[0097] S401, obtaining message keywords according to the message content information;

[0098] S402, obtaining the frequency occurrence probability of the message keywords;

[0099] S403: Obtain message activity based on the word frequency probability and response activity;

[0100] S404: Obtain a real-time authority index according to the message activity and type information correlation.

[0101] As described in steps S401-S404 above, the present invention effectively solves the problem of complex messages and difficulty identifying key points in traditional management by extracting message keywords based on message content. In traditional instant messaging groups, faced with a massive amount of messages, managers and members often spend a considerable amount of time and effort sifting through useful information. However, in this step, using keyword extraction algorithms in natural language processing technology, such as those based on TF-IDF, combined with part-of-speech tagging, it is possible to accurately extract the core points of a message. For example, in a software development project group, by extracting keywords such as "code vulnerability," managers can quickly grasp the message's main point, and members can quickly locate content relevant to their work, significantly improving information acquisition efficiency. Next, step S402, building on step S401, obtains the word frequency probability of message keywords. Traditional management methods struggle to quantify the importance of keywords in messages and cannot distinguish the value differences between different messages. This step calculates the word frequency probability by counting the number of keyword occurrences within a certain range and dividing it by the total number of words. For example, in an academic research group, for frequently appearing professional terms, by calculating the probability of word frequency, its popularity and importance in the current research discussion can be clarified. By dynamically adjusting the statistical range, it can more accurately reflect the current discussion hotspots, providing a key quantitative basis for objectively evaluating the value of messages. Then, step S403 combines the probability of word frequency and response activity to obtain message activity. The traditional message evaluation method is one-sided, focusing only on the content of the message or the number of times members speak, and it is difficult to fully measure the value of the message. In this step, a weighted calculation method is adopted to set weights for the probability of word frequency and response activity according to the characteristics of the group. For example, in a product development group, if a message contains high-frequency keywords about the core functions of a new product and members have high response activity, the message activity obtained through weighted calculation is high, which can more accurately reflect its actual influence and value in the group, overcoming the limitations of traditional evaluation methods. Finally, step S404 obtains a real-time authority index based on the message activity and the type information correlation, obtains a first correlation coefficient based on the message activity, obtains a second correlation coefficient based on the type information correlation, and obtains the cosine similarity with the message keyword vector, where the calculation formula is:

[0102] ;

[0103] in, Indicates the real-time permission index, Indicates message activity. represents the first correlation coefficient, Indicates the type information association, represents the second correlation coefficient, Represents the cosine similarity with the message keyword vector;

[0104] Calculating the real-time authority index involves obtaining cosine similarity with the message keyword vector. Cosine similarity measures the degree of directional similarity between two vectors, ranging from -1 to 1. Values closer to 1 indicate more similar directions; values closer to -1 indicate more opposite directions; and a value of 0 indicates that the two vectors are orthogonal (perpendicular to each other, meaning they are unrelated). In this scenario, calculating cosine similarity with the message keyword vector measures the semantic relevance of different messages to the group topic or other relevant factors, thereby more accurately determining the real-time authority index. Suppose that in an instant messaging group for a software development project, the core topic of discussion is "optimizing user login functionality." The keyword vector C for this topic is extracted, primarily containing keywords such as "user login," "optimization," and "function." These keywords are then converted to vector representations using a word embedding model (such as Word2Vec), resulting in a multidimensional vector. Member A posts a message titled "Fix the user login page lag issue and improve login speed." The keyword vector F for this message is also extracted, containing keywords such as "user login," "lag fix," and "login speed improvement," and converted to a vector. Then use the cosine similarity formula: To calculate the cosine similarity of vectors C and F. C·F is the dot product of two vectors, reflecting the sum of their products in all dimensions; and are the modulos (lengths) of vectors C and F, respectively, , ( is the dimension of the vector). Assume that after calculation, the cosine similarity is 0.8, which indicates that the message of member A has a high semantic similarity with the topic of "Optimizing the User Login Function" discussed in the current group. Combined with the formula If the message activity x is 0.7, the first correlation coefficient G is 0.6, the type information correlation y is 0.7, and the second correlation coefficient H is 0.5, then the real-time permission index A = 0.7 × 0.6 × 0.8 + 0.7 × 0.5 = 0.336 + 0.35 = 0.686. Due to the high similarity between this message and the topic, the high real-time permission index indicates that the message is important to the group discussion. Member A may therefore be granted more permissions, such as code modification permissions, which can better promote the optimization of related features in the project. This fundamentally solves the problem of irrational traditional permission allocation. Traditional permission allocation is often based on factors such as fixed member status and position, ignoring members' actual contributions and behavior within the group. In this step, the same weighted calculation method is used to assign weights to message activity and type information correlation to derive the real-time permission index. In open source project development groups, members who frequently post highly active messages and share attachments that are highly relevant to the project theme will receive a higher real-time permission index, granting them more permissions, such as code submission and document editing. The system also dynamically adjusts member permissions based on the real-time permission index to ensure permissions match member performance and fully motivate them.

[0105] In one embodiment, the step of obtaining a message authority level according to a real-time authority index corresponding to the member interaction information, and obtaining a message prediction value according to the message authority level includes:

[0106] S501, obtaining the corresponding number of participating members according to the member interaction information;

[0107] S502: Obtain the total number of group members;

[0108] S503: Obtaining a message authority level based on the number of participating members, the total number of participating members, and a real-time authority index;

[0109] S504: Acquire information content density according to the member interaction information;

[0110] S505: Obtaining a predicted message reading value based on the content density of the information;

[0111] S506: Obtain a message prediction value according to the predicted message reading value.

[0112] As described in steps S501-S506 above, the present invention provides key data for measuring the impact of a message by obtaining the number of participating members corresponding to member interaction information. In traditional instant messaging group management, it is difficult to determine the actual number of participants in each message discussion, resulting in a lack of basis for determining the popularity and scope of a message's influence. In this step, the system uses a message-member ID correspondence table to monitor new messages and replies in real time, accurately counting the number of participating member IDs. Just like in a corporate strategic planning discussion group, this method allows for intuitive understanding of the number of participants in a new market expansion message and, therefore, to determine its level of attention. Secondly, step S502 obtains the total number of group members, which is crucial for calculating message permission levels. Under traditional management methods, group member statistics can be untimely or inaccurate, hindering the accurate assessment of a message's influence. In this step, the system monitors group membership changes in real time, promptly updating the total number of group members, whether new members join or existing members leave. For example, in an online learning community, the number of members changes as the course progresses. The system ensures data accuracy by interacting with the group management backend, providing reliable denominator data for the scientific calculation of message permission levels, making the evaluation results more objective and comparable. Then, in step S503 (which should be "deriving the message permission level based on the number of participating members, the total number of members, and the real-time permission index"), multiple key factors are integrated to determine the message permission level, overcoming the one-sidedness of traditional single-factor evaluations. Traditional message permission assessments often rely on the identity of the message sender or simple content, failing to fully reflect the importance of the message. In this step, calculation models such as weighted average are used to assign appropriate weights to the proportion of participating members to the total number of members and the real-time permission index (weights can be adjusted based on the nature of the group and management needs). For example, in a software development project group, a message initiated by a core developer (with a high real-time permission index) and participated in by numerous members discussing key feature optimization would be calculated to have a relatively high message permission level. This helps project managers prioritize important messages, allocate resources appropriately, and ensure smooth project progress. Next, in step S504, the content density of the message is derived based on member interaction information, helping to measure the richness and value of the message content. In traditional management, it's difficult to accurately judge the value of message content, and messages with rich but unobtrusive content are easily overlooked. However, in this step, text analysis technology is used to calculate the content density of the message by counting the vocabulary and key information in the message. For example, in a financial investment group, a message analyzing market trends and containing a large amount of data and professional opinions has a high content density and can provide investors with important decision-making information. Then, in step S505, the predicted message reading value is obtained based on the content density of the message.While information content density only measures the content of the message itself, the predicted message reading value further considers the potential benefits members may gain from reading the message. By analyzing information content density and combining it with the group's business goals and member needs, the potential value of the message to members in terms of knowledge acquisition, decision support, and other aspects can be assessed. For example, in an e-commerce operations group, a message with high content density about a new marketing strategy is assessed to have a high predicted message reading value, as it can provide operations staff with new ideas and methods, helping to improve business results. Finally, in step S506, the predicted message reading value is determined based on the predicted message reading value. This message prediction value integrates the analysis results of the previous steps and more comprehensively reflects the potential impact and value of the message within the group. The predicted message reading value is quantified and combined with other relevant factors (such as the timeliness of the message) to obtain the message prediction value. In a product development group, the message prediction value determined based on the predicted message reading value can help managers plan resources and schedule work in advance, giving more attention and support to messages that may promote product development progress.

[0113] In one embodiment, the step of obtaining a resource allocation value according to the message prediction value includes:

[0114] S601, obtaining a message differentiated transmission control value and a resource requirement value according to the message prediction value;

[0115] S602: Obtain a first weight coefficient according to the message differentiated transmission control value;

[0116] S603: Obtain a second weight coefficient according to the resource demand value;

[0117] S604: Obtain a resource allocation value according to the message differentiated transmission control value, the first allocation weight coefficient, the resource requirement value, and the second allocation weight coefficient.

[0118] As described in steps S601-S604 above, the present invention obtains message differential transmission control values and resource requirements based on the message prediction value. Traditional instant messaging groups use a uniform, indiscriminate approach for message transmission and resource allocation, potentially leading to the loss of important messages due to transmission delays and the excessive use of resources by low-value messages. In S601, the system determines the message differential transmission control value based on pre-set rules and algorithms, taking into account the message prediction value, message type, and real-time network conditions. It also determines the resource requirement based on factors such as the message data volume and processing complexity. For example, in a financial transaction group, if a message involving a large transaction decision has a high prediction value, the system will increase its differential transmission control value to ensure priority transmission. It will also determine a higher resource requirement based on its data processing requirements to avoid transaction errors and ensure smooth business operations. Next, in step S602, a first weight coefficient is obtained based on the message differential transmission control value. Traditional resource allocation methods struggle to quantify the impact of message transmission importance on resource allocation, resulting in a lack of scientific basis for resource allocation. In this step, a weight calculation model is established and normalized to map the message's differentiated transmission control value to a reasonable range (e.g., 0-1) to obtain a first weight coefficient. For example, a linear normalization formula can be used to calculate the value based on the maximum value of the message's differentiated transmission control value within a certain time period. This allows the transmission importance of different messages to be converted into comparable weight coefficients. In large enterprise project communication groups, messages related to key project nodes and with high differentiated transmission control values will receive a higher first weight coefficient, receiving more weight in resource allocation and promoting smooth project progress. Then, in step S603, a second weight coefficient is obtained based on the resource demand value. Traditional resource allocation often ignores the differences in resource requirements of messages, resulting in a mismatch between resource allocation and actual demand. S603 converts the resource demand value into the second weight coefficient using a specific algorithm. For example, resource demand values are graded, with different weight coefficients assigned to low, medium, and high levels, such as 0.3, 0.6, and 0.9, and fine-tuned for resource demand values in the intermediate range. Taking an online education live broadcast group as an example, the resource demand value of live course-related messages is high, and the second weight coefficient obtained is also high, thereby giving priority to the allocation of resources such as server bandwidth and storage space, improving the user learning experience and making resource allocation more in line with actual needs. Finally, step S604 obtains the resource allocation value based on the message differentiated transmission control value, the first allocation weight coefficient, the resource demand value, and the second allocation weight coefficient. The calculation formula is:

[0119] ;

[0120] in, Indicates the resource allocation value, represents the first allocation weight coefficient, Indicates the message differential transmission control value, represents the second allocation weight coefficient, represents the resource demand value. In the optimization of instant messaging group management, this formula comprehensively considers multiple key factors to achieve precise resource allocation. The resource demand value L is determined based on factors such as the message data volume and processing complexity, reflecting the message's actual demand for various resources. For example, in a video production project group, messages involving the transmission of high-definition video footage and special effects processing have high resource demands due to their large data volume and complex processing; simple text messages, on the other hand, have relatively low resource demands. The second allocation weight coefficient b is derived from the resource demand value using a specific algorithm to reflect the impact of different resource demand values on resource allocation. For example, resource demand values can be graded, with lower weight coefficients corresponding to low demand levels and higher weight coefficients corresponding to high demand levels. For example, the weight coefficients for low demand levels can be set to 0.3, medium demand levels to 0.6, and high demand levels to 0.9. The weight coefficients can then be fine-tuned for resource demand values in the middle range to better align with actual needs. When calculating resource allocation values, the message differential transmission control value (E) is determined based on preset rules and algorithms, taking into account the message's predicted value, message type, and real-time network conditions. It determines the message transmission priority. The first allocation weight coefficient (a) is derived from the message differential transmission control value. By establishing a weight calculation model and applying a normalization method, it is mapped to a reasonable range (e.g., 0-1). Substituting these factors into the formula, the resource allocation value is calculated. For example, during an e-commerce promotion, a message in an instant messaging group about an inventory update for a limited-time flash sale product has a high predicted value, indicating it is critical to the business. The system determines its message differential transmission control value (E) to be 0.9, and the calculated first allocation weight coefficient (a) is 0.8. Because this message involves real-time updates of a large amount of inventory data, the data volume is large and the processing requirements are high, the resource demand value (L) is 0.8, and the corresponding second allocation weight coefficient (b) is 0.7. According to the formula Q = a * E + b * L, the resource allocation value Q = 0.8 × 0.9 + 0.7 × 0.8 = 0.72 + 0.56 = 1.28. A higher resource allocation value indicates that this message should be prioritized in resource allocation. The system will allocate more network bandwidth, server processing power, and other resources to it, ensuring that inventory information is transmitted to relevant personnel in a timely and accurate manner, ensuring the smooth execution of the limited-time flash sale event, and avoiding business issues caused by message transmission delays or insufficient resources. Traditional resource allocation methods fail to comprehensively consider the various characteristics of messages, which can easily lead to resource waste or insufficient resources for critical messages. In S604, a weighted summation method is used, such as resource allocation value = message differentiated transmission control value × first weight coefficient + resource demand value × second weight coefficient, to organically combine multiple key factors. The system also monitors changes in message parameters in real time and dynamically adjusts resource allocation values.For example, in an instant messaging group for e-commerce promotions, messages related to real-time information about promotions have high message prediction values, high differentiated message transmission control values, and high resource demand values. Combined with the two weight coefficients, the resource allocation value calculated is high, and thus priority is given to resource allocation such as network bandwidth and server processing power, ensuring the smooth progress of the promotion.

[0121] like Figure 2 As shown, the present invention also provides an instant messaging group management optimization system, comprising:

[0122] A first acquisition module 1 is configured to acquire a set of interaction data of group members, and acquire a plurality of member interaction information based on the interaction data set, wherein the member interaction information includes message content information and attachment type information;

[0123] The second acquisition module 2 is used to obtain the response time interval with other members according to the message content information, and obtain the response activity according to the response time interval;

[0124] The third acquisition module 3 is used to obtain the correlation degree of type information with other members according to the attachment type information;

[0125] A fourth acquisition module 4 is configured to acquire a real-time permission index based on the response activity and the type information correlation;

[0126] A fifth acquisition module 5 is configured to acquire a message authority level according to the real-time authority index corresponding to the member interaction information, and acquire a message prediction value according to the message authority level;

[0127] The control module 6 is configured to obtain a resource allocation value according to the message prediction value, and to control member interaction information based on the resource allocation value.

[0128] In one embodiment, the second acquisition module 2 includes:

[0129] A first acquiring unit, configured to acquire multiple conversation threads according to the message content information;

[0130] A second acquiring unit, configured to acquire a timestamp according to the conversation thread;

[0131] A third acquiring unit is configured to generate a time series sorted list of each message in the conversation thread according to the timestamp mark;

[0132] a fourth acquiring unit, configured to acquire time interval differences between adjacent messages according to the time series sorted list, and generate a response time interval set;

[0133] A fifth acquiring unit is configured to acquire a response activity value according to the response time interval set.

[0134] In one embodiment, the third acquisition module 3 includes:

[0135] a sixth acquiring unit, configured to acquire a plurality of group classification labels according to the attachment type information;

[0136] a seventh acquiring unit, configured to establish a group subject-attachment type mapping table according to the plurality of group classification tags;

[0137] An eighth acquiring unit, configured to acquire, according to the group topic-attachment type mapping table, the information sending frequency of the same type of attachments in the current session period;

[0138] The ninth acquiring unit is configured to acquire the type information association degree according to the information sending frequency.

[0139] In one embodiment, the fourth acquisition module 4 includes:

[0140] a tenth acquiring unit, configured to acquire message keywords according to the message content information;

[0141] An eleventh obtaining unit, configured to obtain a word frequency occurrence probability of a message keyword;

[0142] a twelfth acquiring unit, configured to acquire message activity according to the word frequency occurrence probability and the response activity;

[0143] The thirteenth obtaining unit is used to obtain a real-time authority index according to the message activity and the correlation between the type information.

[0144] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, value library or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct RAM bus dynamic RAM (DRDRAM), and RAM bus dynamic RAM (RDRAM).

[0145] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0146] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent results or equivalent process transformations made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for optimizing instant messaging group management, characterized in that: include: Acquire a group member interaction data set, and acquire a plurality of member interaction information according to the interaction data set, wherein the member interaction information includes message content information and attachment type information; Obtaining a response time interval with other members based on the message content information, and obtaining a response activity based on the response time interval; Obtaining a degree of association with other members' type information based on the attachment type information; Obtaining message keywords according to the message content information; Get the frequency probability of the message keywords; Obtaining message activity based on the word frequency probability and response activity; Obtaining a real-time authority index based on the message activity and type information correlation; Acquire the corresponding number of participating members according to the member interaction information; Get the total number of group members; Obtaining the message permission level based on the number of participating members, the total number of participating members, and the real-time permission index; Acquiring content density of information according to the member interaction information; Obtaining a predicted message reading value based on the content density of the information; Obtaining a message prediction value according to the predicted message reading value; Obtaining a message differentiated transmission control value and a resource requirement value according to the message prediction value; Obtaining a first weight coefficient according to the message differentiated transmission control value; Obtaining a second weight coefficient according to the resource demand value; Acquire a resource allocation value according to the message differentiated transmission control value, the first allocation weight coefficient, the resource requirement value, and the second allocation weight coefficient; Regulate member interaction information based on resource allocation values.

2. The instant messaging group management optimization method according to claim 1, characterized in that: The step of obtaining the response time interval with other members according to the message content information, and obtaining the response activity according to the response time interval includes: Acquire multiple conversation threads according to the message content information; Obtaining a timestamp according to the conversation thread; Generate a time series sorted list of each message in the conversation thread according to the timestamp mark; Obtaining time interval differences between adjacent messages according to the time series sorted list, and generating a response time interval set; A response activity value is obtained according to the response time interval set.

3. The instant messaging group management optimization method according to claim 1, characterized in that: The step of obtaining the degree of association with type information of other members according to the attachment type information includes: Acquire multiple group classification labels according to the attachment type information; Establishing a group topic-attachment type mapping table according to the plurality of group classification tags; Obtaining the frequency of information transmission of similar attachments within the current session period according to the group topic-attachment type mapping table; The type information association is obtained according to the information sending frequency.

4. An instant messaging group management optimization system, characterized in that: include: a first acquisition module, configured to acquire an interaction data set of group members, and acquire a plurality of member interaction information according to the interaction data set, wherein the member interaction information includes message content information and attachment type information; A second acquisition module is used to acquire a response time interval with other members according to the message content information, and acquire a response activity according to the response time interval; A third acquisition module is used to acquire the degree of association of type information with other members based on the attachment type information; A fourth acquisition module, configured to acquire message keywords based on the message content information; Get the frequency probability of the message keywords; Obtaining message activity based on the word frequency probability and response activity; Obtaining a real-time authority index based on the message activity and type information correlation; A fifth acquisition module, configured to acquire the corresponding number of participating members according to the member interaction information; Get the total number of group members; Obtaining the message permission level based on the number of participating members, the total number of participating members, and the real-time permission index; Acquiring content density of information according to the member interaction information; Obtaining a predicted message reading value based on the content density of the information; Obtaining a message prediction value according to the predicted message reading value; A control module, configured to obtain a message differentiated transmission control value and a resource requirement value according to the message prediction value; Obtaining a first weight coefficient according to the message differentiated transmission control value; Obtaining a second weight coefficient according to the resource demand value; Acquire a resource allocation value according to the message differentiated transmission control value, the first allocation weight coefficient, the resource requirement value, and the second allocation weight coefficient; Regulate member interaction information based on resource allocation values.

5. The instant messaging group management optimization system according to claim 4, characterized in that: The second acquisition module includes: A first acquiring unit, configured to acquire multiple conversation threads according to the message content information; A second acquiring unit, configured to acquire a timestamp according to the conversation thread; A third acquiring unit is configured to generate a time series sorted list of each message in the conversation thread according to the timestamp mark; a fourth acquiring unit, configured to acquire time interval differences between adjacent messages according to the time series sorted list, and generate a response time interval set; A fifth acquiring unit is configured to acquire a response activity value according to the response time interval set.

6. The instant messaging group management optimization system according to claim 4, characterized in that: The third acquisition module includes: a sixth acquiring unit, configured to acquire a plurality of group classification labels according to the attachment type information; a seventh acquiring unit, configured to establish a group subject-attachment type mapping table according to the plurality of group classification tags; An eighth acquiring unit, configured to acquire, according to the group topic-attachment type mapping table, the information sending frequency of the same type of attachments in the current session period; The ninth acquiring unit is configured to acquire the type information association degree according to the information sending frequency.

7. The instant messaging group management optimization system according to claim 4, characterized in that: The fourth acquisition module includes: a tenth acquiring unit, configured to acquire message keywords according to the message content information; An eleventh obtaining unit, configured to obtain a word frequency occurrence probability of a message keyword; a twelfth acquiring unit, configured to acquire message activity according to the word frequency occurrence probability and the response activity; The thirteenth obtaining unit is used to obtain a real-time authority index according to the message activity and the correlation between the type information.

Images