Community operation evaluation determination method, apparatus and device, and medium
By acquiring community communication records, extracting semantic data, and performing clustering assessments, the problems of dynamic monitoring and public opinion risk identification in community operation evaluation were solved, enabling more accurate and comprehensive community operation evaluation and improving business conversion rates.
Patent Information
- Application Number
- CN202511351281.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-19
- Publication Date
- 2026-01-16
AI Technical Summary
Existing technologies make it difficult to dynamically monitor and quantitatively score community operations, and fail to detect community public opinion risks in a timely manner, which affects corporate image and business conversion.
By acquiring the current communication records of the community, extracting semantic data, determining the business interest of ordinary users and the communication quality and public opinion tags of active users, clustering and evaluating the community based on this data, and dynamically adjusting the evaluation results in conjunction with public opinion tags.
It improves the accuracy and comprehensiveness of community operation evaluation, enabling timely identification of negative public opinion trends, intervention measures, and improved business conversion rates.
Smart Images

Figure CN121351818A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of community operation evaluation determination and the field of financial technology, and in particular to a community operation evaluation determination method, device, equipment and medium. BACKGROUND
[0002] At present, in the process of enterprise community operation, in addition to administrators, Key Opinion Leaders (KOLs) or other active users among users have a strong operation and driving effect on ordinary users. This effect may be positive or negative. Real-time monitoring of the overall operation of the community through conversation archives helps to reduce community risks and improve the conversion willingness of community customers.
[0003] In related technologies, community communication records are mainly analyzed manually to determine the overall operation of the community. This method is inefficient and difficult to achieve dynamic monitoring and quantitative scoring of multiple community operation evaluations. In addition, related technologies usually do not consider the public opinion risk of the community, cannot reasonably distinguish between active users and ordinary users, and cannot determine whether the community as a whole is satisfied with the current policy in a timely manner, thereby affecting the enterprise image and business conversion. SUMMARY
[0004] Embodiments of the present application provide a community operation evaluation determination method, device, equipment and medium to solve the problems of related technologies that are difficult to achieve dynamic monitoring and quantitative scoring of community operation, and cannot detect community public opinion risks in a timely manner.
[0005] In a first aspect, the present disclosure provides a community operation evaluation determination method, comprising: obtaining current communication records of multiple communities; based on the data type of the current communication records, performing data extraction on the current communication records to determine community semantic data; based on the community semantic data, determining the business interest degree of ordinary users and the communication quality and public opinion label of active users in the community; based on the business interest degree and the communication quality, clustering the multiple communities to determine a community clustering result; based on the community clustering result and the public opinion label, determining a current operation evaluation result of the multiple communities.
[0006] In a second aspect, the present disclosure provides a community operation evaluation determination device, comprising: a data acquisition module configured to obtain current communication records of multiple communities; a semantic extraction module configured to, based on the data type of the current communication records, perform data extraction on the current communication records to determine community semantic data; The semantic processing module is configured to determine the business interest degree of the ordinary users in the community, and the communication quality and the public opinion label of the active users based on the community semantic data; The community clustering module is configured to cluster the multiple communities based on the business interest degree and the communication quality, and determine a community clustering result. The operation evaluation module is configured to determine a current operation evaluation result of the multiple communities based on the community clustering result and the public opinion label.
[0007] In a third aspect, the present disclosure provides a computer device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the community operation evaluation determination method described above when executing the computer program. In a fourth aspect, the present disclosure provides a computer readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement the community operation evaluation determination method described above.
[0008] In the scheme implemented by the community operation evaluation determination method, the device, the equipment and the medium described above, the current communication records of the multiple communities are obtained; the data extraction is performed on the current communication records based on the data types of the current communication records, and the community semantic data is determined; the business interest degree of the ordinary users in the community, and the communication quality and the public opinion label of the active users are determined based on the community semantic data; the multiple communities are clustered based on the business interest degree and the communication quality, and a community clustering result is determined; and a current operation evaluation result of the multiple communities is determined based on the community clustering result and the public opinion label. The method of the present disclosure comprehensively evaluates the actual operation of the community from multiple dimensions by determining the business interest degree of the ordinary users, and the communication quality and the public opinion label of the active users, improves the accuracy and comprehensiveness of the current operation evaluation result, realizes the dynamic evaluation of the operation of the community by clustering the community, further improves the accuracy and comprehensiveness of the operation evaluation of the community, and can also judge the current public opinion risk of the community according to the public opinion label, so as to help the operation personnel to identify the potential negative public opinion trend in time and take intervention measures, and to remind the operation personnel to take corresponding incentive measures in time in the positive public opinion, and improve the business conversion rate. BRIEF DESCRIPTION OF DRAWINGS
[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative labor.
[0010] Figure 1 is a system architecture diagram of a community operation evaluation determination method in an embodiment of the present application; Figure 2 is a flowchart of a community operation evaluation determination method in an embodiment of the present application; Figure 3 is another flowchart of a community operation evaluation determination method in an embodiment of the present application; Figure 4 is another flowchart of a community operation evaluation determination method in an embodiment of the present application; Figure 5 is another flowchart of a community operation evaluation determination method in an embodiment of the present application; Figure 6 is another flowchart of a community operation evaluation determination method in an embodiment of the present application; Figure 7 is another flowchart of a community operation evaluation determination method in an embodiment of the present application; Figure 8 is a principle block diagram of a community operation evaluation determination device in an embodiment of the present application; Figure 9 is a principle block diagram of a computer device in an embodiment of the present application. DETAILED DESCRIPTION
[0011] The community operation evaluation determination method provided by the embodiments of the present application can be applied in an application environment such as Figure 1 , including a client and a server as shown in Figure 1 , the client and the server communicate through a network, such as sending the current communication record in the present disclosure to the server to make the server execute the community operation evaluation determination method of the present disclosure. Among them, the client, also known as the user end, is a program that provides local services for customers corresponding to the server. The client can be installed on various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, but is not limited to them. The server can be realized by an independent server or a server cluster composed of multiple servers.
[0012] In an embodiment, as shown in Figure 2 , a community operation evaluation determination method is provided, which is taken as an example to illustrate the server in Figure 1 , including the following steps: S101, obtaining current communication records of multiple communities; S102, based on the data type of the current communication record, performing data extraction on the current communication record to determine community semantic data; S103, based on the community semantic data, determining the business interest degree of ordinary users in the community, and the communication quality and public opinion label of active users; S104, based on the business interest degree and the communication quality, clustering the multiple communities to determine a community clustering result; S105, determining current operation evaluation results of the plurality of communities based on the community clustering results and the public opinion labels.
[0013] As an example, in step S101, the current communication records of the plurality of communities can be obtained by querying a database storing the current communication records, or receiving data uploaded by local devices corresponding to the current communication records, and the disclosure does not limit the manner of obtaining the current communication records.
[0014] It should be understood that the current communication records can be text information, audio information, image information, etc., and the disclosure does not limit the form of the communication records.
[0015] As an example, in step S102, since the current communication records can include text, audio, or image, etc., different processing methods need to be taken for different data types to extract community semantic information in the current communication records. For example, when the data type is text type, natural language processing technology can be used to process the text, such as word segmentation, sentiment analysis, topic modeling, etc., to extract keywords and semantic features and determine community semantic data.
[0016] As an example, in step S103, the activity level of each user in the community can be determined according to the extracted community semantic data, for example, by analyzing the behavior characteristics of the user, such as the frequency of speech, the number of questions and answers, etc., to quantify the activity of the user, thereby classifying the users in the community to determine ordinary users and active users. Further, according to the classification results of the users, different user groups are implemented with targeted evaluation strategies. For ordinary users, the business interest degree shown in the community is analyzed, and for active users, since they have high influence and participation in the community, the communication quality and public opinion label of the active users can be determined according to the community semantic information to judge whether the speech is positive or not, whether there is potential risk or negative emotional tendency, thereby realizing personalized evaluation of community operation.
[0017] The public opinion label is used to indicate the speech tendency, which can include positive, neutral, negative, etc. The public opinion label can be generated by analyzing the semantic features of the active users in the communication, such as keywords, emotional tendency, and topic guiding direction, thereby providing risk warning and quality evaluation basis for community operation.
[0018] As an example, in step S104, based on the business interest degree and the communication quality, clustering of the plurality of communities is realized, thereby dynamically stratifying the plurality of communities to avoid evaluation deviation caused by uniform evaluation standards.
[0019] In other words, by clustering multiple communities, we can analyze the group characteristics of these communities and dynamically determine their operational levels based on these characteristics. This avoids the limitations of using fixed preset thresholds to determine operational assessments and further improves the accuracy and adaptability of the scoring.
[0020] It should be understood that this disclosure does not limit the clustering method used when clustering multiple communities, such as K-means clustering, hierarchical clustering, or density clustering.
[0021] As an example, in step S105, multiple communities can be stratified based on the community clustering results. For instance, communities with high business interest and communication quality can be classified as high-value communities, while communities with moderate business interest and communication quality can be classified as medium-value communities. The stratification results are then integrated with public opinion tags to achieve a comprehensive evaluation of community operations and determine the current operational evaluation result. For example, if a community is determined to be in the high-value community layer based on the clustering results and the public opinion tags are positive, then the current operational evaluation result of that community can be judged as excellent. This indicates that the community not only has high business value but also a healthy and positive community atmosphere, which is conducive to long-term operation and user retention.
[0022] In summary, the method disclosed herein includes: acquiring current communication records from multiple communities; extracting data from the current communication records based on their data type to determine community semantic data; determining the business interest level of ordinary users in the communities, as well as the communication quality and public opinion tags of active users, based on the community semantic data; clustering multiple communities based on business interest level and communication quality to determine the community clustering results; and determining the current operational evaluation results of multiple communities based on the community clustering results and public opinion tags. This method, by determining the business interest level of ordinary users, as well as the communication quality and public opinion tags of active users, comprehensively evaluates the actual operation of communities from multiple dimensions, improving the accuracy and comprehensiveness of the current operational evaluation results. By clustering communities, it achieves dynamic evaluation of community operation, further improving the accuracy and comprehensiveness of community operational evaluation. Simultaneously, it can also judge the current public opinion risks of the community based on public opinion tags, thereby helping operators to promptly identify potential negative public opinion trends and take intervention measures, and promptly reminding operators to take corresponding incentive measures in the event of positive public opinion, thereby improving business conversion rates.
[0023] In one embodiment, step S102, which involves extracting data from the current communication record based on its data type to determine community semantic data, may specifically include: When the data type is an image, perform character recognition and / or image recognition on the current communication record to determine the community semantic data; When the data type is voice, the current communication record is converted into audio, generating the corresponding text data, and features are extracted from the text data to determine the community semantic data. When the data type is text, feature extraction is performed on the current communication record to determine the community semantic data.
[0024] It should be understood that the image type can be a picture or a video stream composed of pictures, and this disclosure does not limit it.
[0025] For example, when the data type is an image, optical character recognition (OCR) technology can be used to recognize the text in the image and extract key information; image recognition technology can also be used to analyze elements such as graphics, expressions, and colors in the image to obtain community semantic data.
[0026] For example, when the data type is speech, speech content can be converted into text data through speech recognition technology, and then natural language processing technology can be used to perform feature processing such as sentiment analysis and keyword extraction on the text data to further mine community semantic data.
[0027] For example, when the data type is text, natural language processing technology can be used to analyze the text content, extract features such as keywords, sentiment, and topic categories, and then determine the semantic data of the community.
[0028] In one embodiment, such as Figure 3 As shown, step S103, which involves determining the business interest level of ordinary users in the community, as well as the communication quality and sentiment tags of active users, based on community semantic data, includes: S201, based on community semantic data, determines the communication frequency, number of questions asked, and number of answers given by each user in the community; S202, users whose communication frequency is greater than or equal to a preset frequency threshold and whose number of answers is greater than the number of questions are identified as active users, and users in the community other than active users are identified as ordinary users; S203, based on ordinary users and active users, classifies and analyzes community semantic data to determine the degree of business interest with ordinary users, as well as the communication quality and public opinion tags of active users.
[0029] As an example, in step S201, communication records of users in the community can be analyzed based on community semantic data to calculate the communication frequency, number of questions asked, and number of answers given by each user within a certain period. For example, the communication frequency can be determined by counting the number of messages a user sends in a day or week, the number of questions asked can be determined by the number of questions in the message content, and the number of answers given can be determined by the number of times a user answers other people's questions.
[0030] As an example, in step S202, users with high communication frequency can be filtered out based on a preset frequency threshold (e.g., 10 messages per day). At the same time, the condition that the number of answers is greater than the number of questions is combined to determine whether the user is an active user; and the remaining users in the community are classified as ordinary users.
[0031] Among them, active users can be individuals with high influence and interaction frequency in the community, such as administrators, key opinion leaders (KOLs), and users who participate in activities for a long time. This disclosure does not limit this.
[0032] As an example, in step S203, the community semantic data corresponding to ordinary users can be analyzed to extract information such as the topic keywords and interest area tags that they are interested in, thereby determining the business interest of ordinary users; at the same time, the community semantic data corresponding to active users can be analyzed in depth to explore the emotional tendencies of their remarks, the professionalism of their communication content, and their impact on the community atmosphere, thereby determining the communication quality and public opinion tags of active users, so as to achieve refined evaluation of different types of community users.
[0033] In one embodiment, such as Figure 4 As shown, step S203, which involves classifying and analyzing the community semantic data based on ordinary users and active users, to determine the level of business interest with ordinary users, as well as the communication quality and public opinion tags of active users, includes: S301, classify and statistically analyze the community semantic data to determine the first semantic data corresponding to ordinary users and the second semantic data corresponding to active users. The first semantic data includes at least one of the following: number of business inquiries, number of activity participations, and current business volume. The second semantic data includes: answering efficiency, guidance tendency, and answering professionalism. S302, determine business interest level based on at least one of the following: number of business inquiries, number of activity participations, and current business volume; S303 determines communication quality based on response efficiency and / or response professionalism, and determines public opinion tags based on guidance tendency.
[0034] As an example, in step S301, statistical analysis of the community semantic data corresponding to active users can be performed to determine indicators such as the efficiency of answering questions, the tendency to guide, and the professionalism of answering questions. Specifically, the efficiency of answering questions can be measured by statistically analyzing the average response time of active users after a user asks a question; the tendency to guide can be determined by analyzing the emotional tone of the words used by active users in their answers, such as encouraging words or questioning statements, to determine whether their guidance is positive or negative; and the professionalism of answering questions can be determined by semantic analysis of the content of their answers, judging whether they contain professional terminology, logical clarity, and accuracy, and at least one of the above parameters is determined as the first semantic.
[0035] As an example, in step S301, statistical analysis can be performed using the community semantic data corresponding to ordinary users to determine indicators such as the number of business inquiries, the number of activities participated in, and the current business volume of ordinary users. Specifically, the number of business inquiries can be measured by the frequency of questions asked by ordinary users within a specific period; the number of activities participated in is measured by the frequency of interaction among ordinary users in business activities conducted within the community; and the current business volume can be assessed by the number of business operations completed by ordinary users within a specific period (e.g., the number of insurance policies purchased in a quarter), and at least one of the above parameters is determined as the second semantic data.
[0036] As an example, in step S302, a weighted calculation can be performed on the number of business inquiries, the number of activity participations, and the current business volume to obtain a comprehensive score, thereby quantifying the business interest of ordinary users. The weights can be adjusted according to actual business needs. For example, in an insurance community, the weight of the current business volume can be appropriately increased to reflect the importance of the user's actual business conversion. Through this score, highly interested users can be accurately identified, thus providing data support for subsequent differentiated operation strategies. However, it is not limited to this; in addition to a score, business interest can also be expressed as qualitative descriptions such as "high interest," "medium interest," or "low interest."
[0037] As an example, in step S303, the efficiency and professionalism of the response can be quantitatively scored. For instance, higher efficiency results in a higher score, while professionalism is scored based on the accuracy of the content and the use of professional terminology. A weighted average is then used to arrive at a comprehensive score for communication quality. Furthermore, if a user displays strong positive emotions or encouraging language in their response, the user can be labeled with a "positive guidance" tag; conversely, they can be labeled with a "negative guidance" tag.
[0038] In one embodiment, such as Figure 5 As shown, step S103, which involves clustering multiple communities based on business interest and communication quality to determine the community clustering results, includes: S401, determine the initial cluster centers from multiple communities; S402, based on user quality and communication quality, determines the cluster distance between multiple communities and the initial cluster center; S403, based on clustering distance, determines the initial cluster centers of multiple communities and generates initial clusters; S404, based on multiple communities, iteratively clusters the initial clusters until the changes in the clusters after iteration meet the preset change threshold, and determines the result when the iteration stops as the community clustering result.
[0039] As an example, in step S401, the initial cluster centers can be determined by random selection or based on preset rules. For example, the initial cluster centers can be a community with high user quality and communication quality and a community with low user quality and communication quality, thus forming the starting point for clustering.
[0040] As an example, in step S402, the clustering distance can be calculated using a weighted Euclidean distance based on two dimensions: user quality and communication quality. For instance, if the weight of user quality is 0.6 and the weight of communication quality is 0.4, the Euclidean distance calculated by weighting both dimensions is determined as the clustering distance. However, this is not a limitation; Manhattan distance or other similarity measures can also be used to determine the clustering distance, and this disclosure does not restrict this method.
[0041] As an example, in step S403, each community can be assigned to the initial cluster center closest to it based on the calculated clustering distance, thereby generating an initial cluster. This process can employ the minimum distance principle to ensure that each community is classified into the best-matching cluster.
[0042] As an example, in step S404, the iterative clustering process can recalculate the centroid of each cluster and recalculate the distance between each community and the new centroid, thereby redistributing the affiliation clusters until the change in the centroid of the cluster is less than a preset threshold, and the fern situation at the time of stopping the iteration is determined as the community clustering result.
[0043] In one embodiment, such as Figure 6 As shown, in step S104, based on the community clustering results and public opinion tags, the current operational evaluation results of multiple communities are determined, including: S501, based on clustering results, determines the operational levels of multiple communities; S502 integrates operational levels and public opinion tags to determine the current operational assessment results of multiple communities.
[0044] As an example, in step S501, the clustering results can be used to determine the clustering distribution of the community and to classify the operational levels accordingly.
[0045] For example, communities in clusters with high user quality and communication quality are classified as high-operation levels, communities in clusters with high user quality but low communication quality are classified as medium-high-operation levels, and so on to form multiple levels.
[0046] As an example, in step S502, based on the results of the operation level division, a corresponding weight coefficient can be set for each level as the first sub-operation scoring weight. At the same time, based on the distribution characteristics of public opinion tags, each community is assigned a weight based on public opinion operation, which is the second sub-operation scoring weight.
[0047] For example, if the community is determined to be positively guided based on public opinion tags, then the second sub-operation score can be determined to have a higher positive weight. If the community is determined to be negatively guided based on public opinion tags, then the second sub-operation score can be determined to have a lower positive weight or a negative weight.
[0048] For example, based on the operational level classification results, the first sub-operational score weight assigned to communities with high operational levels is a higher positive weight, while the first sub-operational score weight assigned to communities with low operational levels is a lower positive weight.
[0049] Then, the weights of the first sub-operational score and the second sub-operational score are combined by superposition or multiplication to determine the current operation evaluation result.
[0050] Optionally, the operational level and public opinion tags can be directly output as the current operational evaluation results.
[0051] In one embodiment, after determining the current operational evaluation results of multiple communities based on community clustering results and public opinion tags in step S104, multiple communities can be ranked according to the current operational evaluation results; and rewards can be distributed to multiple communities or warning information can be generated based on the ranking and public opinion tags using a preset reward and punishment strategy.
[0052] Specifically, reward points or honorary titles can be awarded to communities with high rankings and positive public opinion labels to increase community enthusiasm and thus help improve the conversion rate of business within the community; while warning information can be generated for communities with low rankings and negative public opinion labels, so as to pay close attention to them or take corresponding intervention measures to effectively reduce the risk of the spread of negative public opinion.
[0053] In one embodiment, such as Figure 7 As shown, after step S104, which determines the current operational evaluation results of multiple communities based on community clustering results and public opinion tags, the following steps are also included: S601, obtain historical operational evaluation results of multiple communities and historical communication records of multiple communities; S602, based on historical operational evaluation results, current operational evaluation results, historical communication records, and current communication records, determines the operational evaluation trends of multiple communities; S603 predicts the future operation of multiple communities based on operational assessment trends.
[0054] As an example, in step S601, the historical operational evaluation results of multiple communities in different periods and the corresponding historical communication records can be obtained by accessing a database that stores relevant information. However, this is not limited to this method, and the above data can also be obtained through other means. This disclosure does not limit this method.
[0055] As an example, in step S602, time series analysis or trend fitting can be used to analyze historical operational evaluation results and communication records, and combine historical communication records with current communication records to determine the operational change trend of the community.
[0056] For example, by comparing and analyzing the changes in historical and target operational evaluation results, and combining the interaction frequency, topic popularity, and the latest developments in current communication records, the activity and operational trend of the community can be assessed, thereby determining the operational score trend of the community.
[0057] As an example, in step S603, based on the prediction results of the operational score trend, regression analysis or machine learning models can be used to quantitatively predict the future operation of the community, thereby providing data support for the adjustment of subsequent community operation strategies.
[0058] In summary, the publicly available method comprehensively evaluates the actual operation of a community from multiple dimensions by determining user quality, communication quality, and public opinion tags, thereby improving the accuracy and comprehensiveness of the target operation assessment results. By introducing public opinion tags, it avoids evaluation biases caused by relying solely on user quality and communication quality. Cluster analysis of multiple communities enables dynamic adjustment of the operation score weights, avoiding the limitations of using fixed weights and further improving the accuracy and adaptability of the target operation assessment results. Based on rankings and public opinion tags, and utilizing pre-set reward and punishment strategies, it effectively incentivizes and manages multiple communities, enhancing overall community activity and positive dissemination effects. Furthermore, by predicting the future operation of multiple communities based on operation score trends, it provides a forward-looking decision-making basis for community operations, further improving the scientific nature and efficiency of community management.
[0059] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0060] In one embodiment, a community operation assessment and determination device is provided, which corresponds one-to-one with the community operation assessment and determination method described in the above embodiments. For example... Figure 8 As shown, the community operation assessment and determination device includes a data acquisition module 701, a semantic extraction module 702, a semantic processing module 703, a community clustering module 704, and an operation assessment module 705. Detailed descriptions of each functional module are as follows: Data acquisition module 701 is used to acquire the current communication records of multiple communities; The semantic extraction module 702 is used to extract data from the current communication record based on the data type of the current communication record to determine the community semantic data; The semantic processing module 703 is used to determine the business interest of ordinary users in the community, as well as the communication quality and public opinion tags of active users, based on community semantic data. The community clustering module 704 is used to cluster multiple communities based on business interest and communication quality, and determine the community clustering results. The Operational Evaluation Module 705 is used to determine the current operational evaluation results of multiple communities based on community clustering results and public opinion tags.
[0061] In one embodiment, the semantic extraction module 702 is further configured to, when the data type is an image type, perform character recognition and / or image recognition on the current communication record to determine community semantic data; When the data type is voice, the current communication record is converted into audio, generating the corresponding text data, and features are extracted from the text data to determine the community semantic data. When the data type is text, feature extraction is performed on the current communication record to determine the community semantic data.
[0062] In one embodiment, the semantic processing module 703 is further configured to determine the communication frequency, number of questions asked, and number of answers given by each user in the community based on community semantic data. Users whose communication frequency is greater than or equal to a preset frequency threshold and whose number of answers is greater than the number of questions are identified as active users, while users in the community other than active users are identified as ordinary users. Based on ordinary users and active users, the semantic data of the community is classified and analyzed to determine the degree of business interest with ordinary users, as well as the communication quality and public opinion tags of active users.
[0063] In one embodiment, the semantic processing module 703 is further configured to classify and statistically analyze the community semantic data to determine the first semantic data corresponding to ordinary users and the second semantic data corresponding to active users. The first semantic data includes at least one of the following: number of business inquiries, number of activity participations, and current business volume. The second semantic data includes: answering efficiency, guidance tendency, and answering professionalism. Determine business interest level based on at least one of the following: number of business inquiries, number of event participations, and current business volume; Based on the efficiency and / or professionalism of the responses, determine the quality of communication, and based on the guiding tendency, determine the public opinion tags.
[0064] In one embodiment, the community clustering module 704 is further configured to determine initial cluster centers from multiple communities; Based on user quality and communication quality, the cluster distance between multiple communities and the initial cluster center is determined; Based on clustering distance, the initial cluster centers of multiple communities are determined, and initial clusters are generated; Based on multiple communities, the initial clusters are iteratively clustered until the changes in the clusters after iteration meet the preset change threshold, and the result when the iteration stops is determined as the community clustering result.
[0065] In one embodiment, the operation evaluation module 705 is further configured to determine the operation level of multiple communities based on the clustering results; By integrating operational levels and public opinion tags, the current operational evaluation results of multiple communities can be determined.
[0066] In one embodiment, the operation evaluation module 705 is further configured to obtain historical operation evaluation results of multiple communities and historical communication records of multiple communities; Based on historical operational evaluation results, current operational evaluation results, historical communication records, and current communication records, the operational evaluation trends of multiple communities are determined; Based on operational assessment trends, predict the future operation of multiple communities.
[0067] This invention provides a community operation evaluation and determination device, comprising: a data acquisition module for acquiring current communication records of multiple communities; a semantic extraction module for extracting data from the current communication records based on their data type to determine community semantic data; a semantic processing module for determining the business interest level of ordinary users in the community, as well as the communication quality and public opinion tags of active users, based on the community semantic data; a community clustering module for clustering multiple communities based on business interest level and communication quality to determine the community clustering results; and an operation evaluation module for determining the current operation evaluation results of multiple communities based on the community clustering results and public opinion tags. This publicly disclosed device comprehensively evaluates the actual operation of a community from multiple dimensions by determining the business interest level of ordinary users, as well as the communication quality and public opinion tags of active users. This improves the accuracy and comprehensiveness of the current operation assessment results. By clustering the community, it achieves dynamic evaluation of the community's operation, further improving the accuracy and comprehensiveness of the community operation assessment. At the same time, it can also judge the current public opinion risks of the community based on public opinion tags, thereby helping operators to identify potential negative public opinion trends in a timely manner and take intervention measures. In the event of positive public opinion, it can promptly remind operators to take corresponding incentive measures to improve business conversion rates.
[0068] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 9 As shown. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database is used for data employed in the community operation assessment and determination method. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the aforementioned community operation assessment and determination method.
[0069] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the aforementioned community operation evaluation method.
[0070] In one embodiment, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described community operation evaluation method.
[0071] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can 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 a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0072] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0073] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A community operation evaluation method characterized by comprising: The method comprises the following steps: obtaining current communication records of a plurality of communities; based on the data type of the current communication records, data extraction is performed on the current communication records to determine community semantic data; based on the community semantic data, the business interest degree of ordinary users in the community, and the communication quality and public opinion label of active users are determined; based on the business interest degree and the communication quality, the plurality of communities are clustered to determine a community clustering result; based on the community clustering result and the public opinion label, a current operation evaluation result of the plurality of communities is determined.
2. The method of claim 1, wherein, The method comprises the following steps: when the data type is an image type, character recognition and / or pattern recognition are performed on the current communication records to determine the community semantic data; when the data type is a voice type, audio conversion is performed on the current communication records to generate text data corresponding to the current communication records, and feature extraction is performed on the text data to determine the community semantic data; when the data type is a text type, feature extraction is performed on the current communication records to determine the community semantic data.
3. The method of claim 1, wherein, The method comprises the following steps: based on the community semantic data, the communication frequency, the number of questions and the number of answers of each user in the community are determined; users with a communication frequency greater than or equal to a preset frequency threshold and a number of answers greater than a number of questions are determined as active users, and users in the community other than the active users are determined as ordinary users; based on the ordinary users and the active users, the community semantic data is classified and analyzed to determine the business interest degree of the ordinary users, and the communication quality and public opinion label of the active users.
4. The method of claim 3, wherein, The method comprises the following steps: the community semantic data is classified and counted to determine first semantic data corresponding to the ordinary users and second semantic data corresponding to the active users, the first semantic data comprising at least one of the number of business consultations, the number of activity participations and the current business volume, and the second semantic data comprising the answer efficiency, the guidance tendency and the answer professional degree; based on at least one of the number of business consultations, the number of activity participations and the current business volume, the business interest degree is determined; based on the answer efficiency and / or the answer professional degree, the communication quality is determined, and based on the guidance tendency, the public opinion label is determined.
5. The method of claim 1, wherein, The method comprises the following steps: an initial clustering center is determined from the plurality of communities; based on the user quality and the communication quality, a clustering distance between the plurality of communities and the initial clustering center is determined; determine initial clustering centers of the multiple community attributions based on the clustering distances, and generate initial clustering clusters; perform iterative clustering on the initial clustering clusters based on the multiple communities until a change of the clustering clusters after iteration meets a preset change threshold, and determine a result when the iteration is stopped as the community clustering result.
6. The method of claim 1, wherein, determine a current operation evaluation result of the multiple communities based on the community clustering result and the public opinion label, including: determine operation levels of the multiple communities based on the clustering result; fuse the operation levels and the public opinion label to determine the current operation evaluation result of the multiple communities.
7. The method of claim 1, wherein, After the determination of the current operation evaluation result of the multiple communities based on the community clustering result and the public opinion label, the method further includes: obtain historical operation evaluation results of the multiple communities and historical communication records of the multiple communities; determine operation evaluation trends of the multiple communities based on the historical operation evaluation results, the current operation evaluation result, the historical communication records, and the current communication records; predict future operations of the multiple communities based on the operation evaluation trends.
8. A community operation evaluation determining apparatus characterized by comprising: The method includes: a data acquisition module configured to obtain current communication records of multiple communities; a semantic extraction module configured to perform data extraction on the current communication records based on data types of the current communication records to determine community semantic data; a semantic processing module configured to determine business interest degrees of ordinary users and communication qualities and public opinion labels of active users in the communities based on the community semantic data; a community clustering module configured to perform clustering on the multiple communities based on the business interest degrees and the communication qualities to determine a community clustering result; an operation evaluation module configured to determine a current operation evaluation result of the multiple communities based on the community clustering result and the public opinion label.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor implements the community operation evaluation determination method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program implements the community operation evaluation determination method according to any one of claims 1 to 7 when executed by the processor.
