Informatization management platform based on public opinion processing flow

Abstract

The invention relates to the technical field of big data and artificial intelligence, in particular to an informatization management platform based on a public opinion processing flow, which comprises a transmissibility data analysis module, a public opinion processing interaction module, a public opinion security situation awareness module and an emergency response and maintenance module. Massive self-media information is integrated and processed, accurate public opinion keyword tags are generated at the same time, and a solid foundation is provided for subsequent public opinion processing; the public opinion task type is automatically and intelligently divided, the task issuing time interval is optimized, and the abnormal public opinion situation is found in time, so that the public opinion processing efficiency and accuracy are remarkably improved; through real-time monitoring and quantitative analysis of the feedback event, an emergency response process is rapidly started, potential risks are effectively reduced, and security and stability of a network space are maintained. The method is used for solving the technical problem of inaccurate public opinion processing analysis in current public opinion management.

Description

Technical Field

[0001] The present invention relates to the technical fields of big data and artificial intelligence, and in particular to an information management platform based on an opinion handling process. Background Art

[0002] Big data refers to data sets that are huge in scale, growing rapidly, and diverse in type. Traditional data processing methods cannot complete the collection, storage, management, and analysis of data within a reasonable time; artificial intelligence technology refers to technologies and methods that simulate and implement human intelligence through machines.

[0003] With the continuous progress of technology and the continuous expansion of application fields, new progress has also been made in the application in the fields of public management and social services. In current opinion management, the following problems still exist: the issuance of opinion tasks lacks scientificity. Traditional opinion management may lack a scientific task issuance mechanism, resulting in overly dense or sparse task issuance, affecting the processing effect; the emergency response is not timely. In the face of urgent opinion events, existing opinion management may be slow to respond and unable to quickly make effective countermeasures, reducing the efficiency and accuracy of dealing with opinion events; the opinion handling strategy lacks flexibility. Traditional opinion management may lack flexible handling strategies and is difficult to adapt to the constantly changing opinion situation. For this reason, the present invention proposes an information management platform based on an opinion handling process. Summary of the Invention

[0004] The purpose of the present invention is to solve the problems in the background art and propose an information management platform based on an opinion handling process.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions: An information management platform based on an opinion handling process, comprising: a communication power data analysis module, an opinion handling interaction module, an opinion security situation perception module, and an emergency response and maintenance module; Communication power data analysis module: Obtain communication power data, perform statistical analysis on it to obtain a comprehensive communication power index, calculate a multi-dimensional influence index according to the comprehensive communication power index, judge whether the multi-dimensional influence index is normal, and generate opinion keyword tags; Opinion handling interaction module: Obtain opinion keyword tags, automatically divide opinion task types according to the opinion keyword tags, set and issue corresponding opinion tasks based on the divided opinion task types, and determine the time interval for the execution of opinion tasks; among them, the opinion handling interaction module includes an opinion task division unit and an opinion task execution unit; Opinion security situation perception module: Perform situation analysis on the completed opinion tasks, judge whether a feedback event needs to be formed according to the situation analysis result, and send the formed feedback event to the emergency response and maintenance module; Emergency Response and Maintenance Module: Conduct real-time monitoring and quantitative analysis on feedback events, and perform processing and control.

[0006] It should be noted that the application object of an information management platform based on the public opinion processing process in the embodiments of the present invention can be public opinion management in the fields of public management and social services. Specifically, it can comprehensively analyze and accurately process the mainstream information data of self-media through big data and artificial intelligence technologies to optimize the public opinion processing process. Through multiple links such as communication power data analysis, public opinion task division and distribution, public opinion security situation perception, and emergency response and maintenance, it can improve the efficiency and accuracy of public opinion processing, meet the urgent needs of organizations such as enterprises for public opinion management, and achieve timely monitoring, effective response, and proper control of public opinion. In summary, the information management platform based on the public opinion processing process has the advantages of being systematic, data-driven, automated and intelligent, and real-time response and early warning, and plays an important role in improving public opinion processing capabilities, optimizing resource allocation, reducing public opinion risks, and providing decision-making support.

[0007] Furthermore, the process by which the communication power data analysis module obtains communication power data and statistically analyzes it to obtain the comprehensive communication power index includes: Collect mainstream self-media information from social media platforms and standardize the information published by self-media; among them, the mainstream self-media information includes title, click-through rate, comment volume, like volume, and repost volume. It can be understood that since the measurement scales among different self-media may be inconsistent, data standardization processing is required to standardize the data. Taking the title, click-through rate, comment volume, like volume, and repost volume of the collected mainstream self-media information as the original data, the Min-Max normalization method is used to process each item of mainstream self-media information respectively, that is, scale each item of original data into the interval [0,1]: standardized data = (original data - minimum value) / (maximum value - minimum value); where the maximum value and minimum value correspond to the original data of each item of mainstream self-media information. Mark the mainstream self-media information that has undergone standardization processing as communication power data , and use a five-tuple to represent the marked communication power data for multi-dimensional representation: ; among them, the five-tuple consists of five elements: title , click-through rate , comment volume , like volume , and repost volume ; Statistically analyze all communication power data in chronological order and integrate to obtain the comprehensive communication power index: Wherein, represents the comprehensive dissemination power index; represents each element of the dissemination power data, that is, ; represents the number of each element of the dissemination power data; represents the weight coefficient of each element in the dissemination power data, which is used to consider the relative importance between different elements; represents the time weight function of each element in the dissemination power data. Since each element changes with time and the importance of each element is different at different time points, a time weight function is introduced for each element, and this time weight function satisfies the normalization condition, that is, the sum at the same time point is equal to 1.

[0008] Furthermore, the process of the dissemination power data analysis module calculating the multi-dimensional influence index, judging and analyzing whether the multi-dimensional influence index is normal, and generating the public opinion keyword tags includes: Calculating and analyzing the multi-dimensional influence index according to the comprehensive dissemination power index: Formula, represents the multi-dimensional influence index; is a preset proportionality coefficient; represents the time window size for calculating the average value, that is, when calculating the multi-dimensional influence index at the previous time point , the data of past time points will be referred to calculate the historical average value, and the purpose is to smooth short-term fluctuations and capture more stable long-term trends; is an index variable, which is used to accumulate the comprehensive dissemination power index of past time points ; is a non-linear effect function, which is evaluated based on the second derivative of, and is expressed as: , where are the first derivative and the second derivative of respectively, representing the change rate and acceleration of; When analyzing the calculation result of the multi-dimensional influence index , all the multi-dimensional influence indexes are sorted in descending order according to the numerical size of the multi-dimensional influence index, and the mean value and the standard deviation of the multi-dimensional influence index are obtained; Combining the mean value and the standard deviation Determine whether the multi-dimensional influence index is within the normal range; Among them, , , in the formula, represents the quantity of the multi-dimensional influence index, represents the index of the multi-dimensional influence index; set the detection limit value of the outlier , and regard the multi-dimensional influence index higher than as an outlier, otherwise, regard it as a normal value. In the formula, represents the selected multiple for defining the outlier; It can be understood that the detected outlier indicates a potential public opinion outbreak point; Collect all abnormal multi-dimensional influence indexes and determine the comprehensive communication power index corresponding to the abnormal multi-dimensional influence index ; ; Collect the text data matching this comprehensive communication power index , including news titles, comments or social media posts, and organize this text data into public opinion information; Use text analysis tools to perform TF-IDF analysis on the associated text and screen for high-frequency words: In the formula, represents the term; represents the document; represents the term appearing in the document ; represents the total number of words in the document ; represents the term appearing in the document and the total number of words in this document , that is, the word frequency; represents the total number of the entire document set ; represents the number of documents containing the term ; represents the inverse document frequency, used to evaluate the general importance of a single word in the entire document set ; Identify high-frequency words with high discrimination for the document set and select this word as the public opinion keyword; Generate corresponding public opinion keyword tags in combination with the public opinion keywords.

[0009] Furthermore, the process by which the public opinion task division unit automatically divides the public opinion task type for the public opinion keyword tags includes: Obtain public opinion keyword tags and set a set of public opinion keyword tags ; According to the public opinion information containing the set of public opinion keyword tags Automatically divide the types of public opinion tasks. Each type of public opinion task includes a forwarding task, a commenting task, and a contribution task. And each time a public opinion task is issued, only one type of public opinion task is allowed to be selected.

[0010] Furthermore, the process in which the public opinion task execution unit sets and issues corresponding public opinion tasks according to the divided types of public opinion tasks and determines the time interval for the execution of public opinion tasks includes: Suppose there is a set of public opinion tasks Needing to be issued; among them, Represents the total number of public opinion tasks; Each public opinion task has a corresponding fixed issuance time ; among them, Represents the public opinion task index; Define the issuance time interval between the current public opinion task And the next public opinion task As ; Set the constraint conditions for the issuance of public opinion tasks: the minimum time interval And the maximum time interval , used to limit the frequency of issuing public opinion tasks and avoid overly dense or sparse issuance of tasks; Assign a corresponding issuance priority to each public opinion task , and determine the issuance order of public opinion tasks according to the priority of public opinion tasks; Initialize the issuance time and set the corresponding initial issuance time For the first public opinion task ; For each subsequent public opinion task , calculate the time interval Between the current public opinion task And the next public opinion task ; According to the constraint conditions of the minimum time interval And the maximum time interval Adjust the time interval ; Calculate the issuance time Of the next public opinion task ; Record the execution results of each public opinion task, specifically including: the type of public opinion task, the corresponding public opinion keyword tags of this type, the issuance time of the public opinion task, and the issuance priority; Integrate and package the recorded information into a public opinion data package for subsequent analysis.

[0011] Furthermore, the public opinion security situation awareness module performs situation analysis on the completed public opinion tasks, and the process of determining whether a feedback event needs to be generated according to the situation analysis results includes: Obtain the public opinion data package, and use the user behavior log to capture the response time, user participation, and public opinion task effect of each public opinion task in the public opinion data package; the response time of the public opinion task refers to the time delay between the release of public opinion and the outbreak of public opinion; the user participation refers to the number of user interactions related to the public opinion task, such as the number of comments and the number of likes; the public opinion task effect refers to the evaluation of the social influence of the public opinion task, such as the scope of dissemination and media exposure; Extract Response time of public opinion tasks , No. Actual user participation in public opinion tasks and The original effect size of the public opinion task ; Perform situation analysis on the extracted public opinion task response time, actual user participation, and original effect value: The response time of public opinion tasks is scaled by power law to obtain , where It represents the response time of the public opinion task after completing the power law scaling. Represents the power law scaling exponent, which is a positive real number greater than 1. It is used to amplify the impact of response time on public opinion tasks. When the response time of a public opinion task is abnormal for a long time, a larger Can significantly improve The value of , emphasizes the negative effects of long delays; Tracking the nonlinear change trend of actual user engagement , where represents the user engagement after nonlinear changes, represents the maximum possible level of user engagement. Setting this upper limit helps to make standardized comparisons. The steepness of the curve controls the tightness of the nonlinear relationship between user participation and public opinion tasks. The value indicates that if the user participation is slightly lower, the steepness of the significant curve will be reduced, otherwise it will be more tolerant; The dynamic range of the original effect value of the public opinion task is compressed to obtain , where It represents the effect value of the public opinion task after dynamic range compression. is the inverse tangent function, Denotes the amplification factor, which is used to enhance the variation range of the original effect value of the public opinion task, and achieve a more delicate resolution effect within a small range. It is particularly suitable for highlighting those events with small effects but requiring attention; Vertically compare the current situation analysis results with the historical public opinion data of similar themes to determine whether a feedback event needs to be formed. If the current situation significantly deviates from the average level or trend line of the historical data, it indicates that there is an anomaly or a situation that requires attention, thus triggering the formation of a feedback event; among them, the historical public opinion data of similar themes refers to the public opinion of similar themes in other regions or industries.

[0012] Furthermore, the process of real-time monitoring, quantitative analysis, processing, and control of the feedback event by the emergency response and maintenance module includes: Obtain the corresponding 、 And ; Real-time monitor the feedback event and assign a unique identifier to each monitored feedback event; Quantitatively analyze the feedback event using the formula: In the formula, Denotes the comprehensive evaluation value of the feedback event; Are weight coefficients, respectively used to adjust 、 And The proportion of the comprehensive evaluation of the feedback event; Is the sensitivity controller, used to control The speed of influence on . Among them, if The value is larger, it means that The influence on Is more intense; Is the sensitivity controller, used to adjust The influence speed. Among them, if The value is larger, it means that The influence on Is more gentle; Is the hyperbolic tangent function, used to map To the interval from -1 to 1; Combine the comprehensive evaluation value Of the feedback event to set the response threshold ; Among them, the response threshold Is used to distinguish feedback events of different emergency levels or importance; When comparing the calculated comprehensive evaluation value Of the feedback event with the preset response threshold : When the comprehensive evaluation value of the feedback event Higher than the response threshold When it is, it indicates that the feedback event has a high level of urgency, with relatively high urgency or importance, and may have a greater impact on society, the organization, or users. It is necessary to immediately initiate the emergency response process and generate a warning prompt. The emergency response process specifically includes: sending a warning prompt to relevant personnel to ensure they can quickly understand the public opinion situation; organizing forces to respond, including allocating resources and deploying personnel; collecting more information to further understand the development trend and potential impact of the public opinion; and adjusting the public opinion handling strategy according to the specific situation to ensure that the public opinion is properly controlled. When the comprehensive evaluation value of the feedback event Is not higher than the response threshold When it is, it indicates that the feedback event has a low level of urgency and has not reached the level that requires an immediate response. Continue to monitor the dynamic changes of the feedback event, and at the same time formulate corresponding preventive measures for the feedback event to reduce potential risks; After the current public opinion control ends, remove the public opinion keyword tags corresponding to the feedback event of the current identifier; among them, the feedback event of the current identifier refers to the feedback event currently being processed.

[0013] Compared with the existing technologies, the advantages of the information management platform and method based on the public opinion handling process provided by the present invention are as follows: 1. The present invention obtains the dissemination power data, statistically analyzes it to obtain the comprehensive dissemination power index, calculates the multi-dimensional influence index according to the comprehensive dissemination power index, judges and analyzes whether the multi-dimensional influence index is normal, and generates public opinion keyword tags, which helps to accurately identify public opinion hotspots and provides strong support for subsequent public opinion handling; 2. The present invention obtains the public opinion keyword tags, automatically divides the public opinion task types according to the public opinion keyword tags, sets and issues corresponding public opinion tasks based on the divided public opinion task types and determines the time interval for the execution of the public opinion tasks, improves the efficiency and accuracy of public opinion handling, reduces the cost of manual intervention, reasonably controls the issuance frequency and order of public opinion tasks, and ensures the continuity and stability of public opinion handling; 3. The present invention conducts a situation analysis on the completed public opinion tasks, judges whether it is necessary to form a feedback event according to the situation analysis result, which helps to accurately evaluate the severity and development trend of the public opinion situation; conducts real-time monitoring and quantitative analysis on the feedback event, and conducts processing and control to ensure that the public opinion is handled in a timely and effective manner.

[0014] In summary, the present invention integrates multiple modules such as dissemination power data analysis, public opinion handling interaction, public opinion security situation perception, and emergency response and maintenance, realizes the full-process information management of public opinion handling, improves the efficiency and accuracy of public opinion handling, and ensures the efficient and stable operation of an information management platform based on the public opinion handling process. Description of the Drawings

[0015] Figure 1 This is a module diagram of an information management platform based on the public opinion processing process proposed by the present invention. Detailed Embodiments

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0017] Refer to Figure 1 , an information management platform based on the public opinion processing process. The system includes a communication power data analysis module, a public opinion processing interaction module, a public opinion security situation perception module, and an emergency response and maintenance module; Communication power data analysis module: Obtain communication power data, perform statistical analysis on it to obtain a comprehensive communication power index, calculate a multi-dimensional influence index based on the comprehensive communication power index, determine whether the multi-dimensional influence index is normal, and generate public opinion keyword tags; Public opinion processing interaction module: Obtain public opinion keyword tags, automatically divide public opinion task types according to the public opinion keyword tags, set and issue corresponding public opinion tasks based on the divided public opinion task types, and determine the time interval for the execution of public opinion tasks; among them, the public opinion processing interaction module includes a public opinion task division unit and a public opinion task execution unit; Public opinion security situation perception module: Perform situation analysis on the completed public opinion tasks, determine whether a feedback event needs to be formed according to the situation analysis result, and send the formed feedback event to the emergency response and maintenance module; Emergency response and maintenance module: Perform real-time monitoring and quantitative analysis on the feedback event, and perform processing and control.

[0018] It should be noted that the application object of an information management platform based on the public opinion processing process in the embodiments of the present invention can be public opinion management in the field of public management and social services. Specifically, it can comprehensively analyze and accurately process the mainstream information data of self-media through big data and artificial intelligence technologies to optimize the public opinion processing process. Through multiple links such as communication power data analysis, public opinion task division and distribution, public opinion security situation perception, and emergency response and maintenance, it can improve the efficiency and accuracy of public opinion processing, meet the urgent needs of organizations such as enterprises for public opinion management, and achieve timely monitoring, effective response, and proper control of public opinion. To sum up, the information management platform based on the public opinion processing process has the advantages of systematization, data-driven, automation and intelligence, and real-time response and early warning, and plays an important role in improving public opinion processing capabilities, optimizing resource allocation, reducing public opinion risks, and providing decision-making support.

[0019] The steps of the communication power data analysis module obtaining communication power data, statistically analyzing it to obtain a comprehensive communication power index, calculating a multi-dimensional influence index based on the comprehensive communication power index, judging whether the multi-dimensional influence index is normal, and generating public opinion keyword tags include: Step 101: Collect mainstream self-media information from social media platforms and standardize the information published by self-media; among them, the mainstream self-media information includes title, click-through rate, comment volume, like volume, and repost volume. In step 101, since the measurement scales among different self-media may be inconsistent, data standardization processing is required to standardize the data. Taking the title, click-through rate, comment volume, like volume, and repost volume of the collected mainstream self-media information as the original data, the Min-Max normalization method is used to process each item of mainstream self-media information respectively, that is, scaling each item of original data into the [0,1] interval: standardized data = (original data - minimum value) / (maximum value - minimum value); where the maximum value and the minimum value correspond to the original data of each item of mainstream self-media information. Step 102: Mark the mainstream self-media information that has been standardized as communication power data , and use a five-tuple to represent the marked communication power data in multiple dimensions: ; where the five-tuple consists of five elements: title , click-through rate , comment volume , like volume , and repost volume ; Step 103: Statistically analyze all communication power data in chronological order and integrate to obtain a comprehensive communication power index: In the formula, Denote the comprehensive dissemination power index; Denote each element of the dissemination power data, i.e., ; Denote the number of each element of the dissemination power data; Denote the weight coefficient of each element in the dissemination power data, which is used to consider the relative importance between different elements; Denote the time weight function of each element in the dissemination power data. Since each element changes over time and the importance of each element is different at different time points, a time weight function is introduced for each element, and this time weight function satisfies the normalization condition, that is, the sum at the same time point is equal to 1; Step 104: Calculate and analyze the multi-dimensional influence index according to the comprehensive dissemination power index: Formula, Denote the multi-dimensional influence index; Is a preset proportionality coefficient; Denote for calculating The time window size of the average value, that is, when calculating the multi-dimensional influence index at the previous time point The multi-dimensional influence index Will refer to the data of past time points to calculate the historical Average value, the purpose is to smooth short-term fluctuations and capture more stable long-term trends; Is an index variable, used to accumulate the comprehensive dissemination power index of past time points For example, if , then when calculating the multi-dimensional influence index at Time, will consider These five time points of Value; Is a non-linear effect function, evaluated based on The second derivative of, expressed as: , where Are respectively The first derivative and the second derivative of, representing The change rate and acceleration of; Step 105: When analyzing the calculation result of the multi-dimensional influence index , sort all the multi-dimensional influence indexes In descending order according to the numerical size of the multi-dimensional influence index, and obtain the mean value And the standard deviation ; Step 106: Combine the mean value And the standard deviation Determine whether the multi-dimensional influence index is within the normal range; Among them, , , in the formula, represents the quantity of the multi-dimensional influence index, represents the index of the multi-dimensional influence index; set the detection limit value of the outlier , and regard the multi-dimensional influence index higher than as an outlier, otherwise, regard it as a normal value. In the formula, represents the selected multiple, which is used to define the outlier; In step 106, the detected outlier indicates a potential public opinion outbreak point; Step 107, collect all abnormal multi-dimensional influence indexes and determine the comprehensive communication power index corresponding to the abnormal multi-dimensional influence index ; ; Step 108, collect the text data matching this comprehensive communication power index , including news titles, comments or social media posts, and organize this text data into public opinion information; Step 109, use a text analysis tool to perform TF-IDF analysis on the associated text and screen for high-frequency words: In the formula, represents the term; represents the document; represents the term appearing in the document ; represents the total number of words in the document ; represents the term appearing in the document and the total number of words in this document , that is, the word frequency; represents the total number of the entire document set ; represents the number of documents containing the term ; represents the inverse document frequency, which is used to evaluate the general importance of a single word in the entire document set ; Step 110, identify the words that are high-frequency and have high discrimination for the document set, and select this word as the public opinion keyword; Step 111, generate the corresponding public opinion keyword label in combination with the public opinion keyword.

[0020] The steps for the public opinion handling interaction module to obtain public opinion keyword tags, automatically classify public opinion task types according to the public opinion keyword tags, set and issue corresponding public opinion tasks based on the classified public opinion task types, and determine the time interval for the execution of public opinion tasks include: Step 201: Obtain public opinion keyword tags and set a public opinion keyword tag set ; Step 202: Automatically classify public opinion task types according to the public opinion information containing the public opinion keyword tag set . Each public opinion task type includes a forwarding task, a commenting task, and a contribution task, and only one public opinion task type is allowed to be selected each time a public opinion task is issued; Step 203: Assume that there is a group of public opinion tasks that need to be issued; where represents the total number of public opinion tasks; Step 204: Each public opinion task has a corresponding fixed issuance time ; where represents the public opinion task index; Step 205: Define the issuance time interval between the current public opinion task and the next public opinion task as ; Step 206: Set the constraint conditions for public opinion task issuance: the minimum time interval and the maximum time interval to limit the issuance frequency of public opinion tasks and avoid overly dense or sparse task issuance; Step 207: Assign a corresponding issuance priority to each public opinion task, and determine the issuance order of public opinion tasks according to the priority of public opinion tasks; Step 208: Initialize the issuance time and set the corresponding initial issuance time for the first public opinion task ; Step 209: For each subsequent public opinion task , calculate the time interval between the current public opinion task and the next public opinion task ; Step 210: Adjust the time interval according to the constraint conditions of the minimum time interval and the maximum time interval ; Step 211: Calculate the issuance time of the next public opinion task ; For example, assume there are three public opinion tasks , and the corresponding initial issuance time intervals are respectively and ; determine the minimum time interval minutes and the maximum time interval minutes of the constraint condition; set the initial issuance time of the first public opinion task to be immediately issued as minutes; calculate and , and determine whether it meets ; if the constraint conditions and are not met, make adjustments until it meets the constraint conditions; use the adjusted issuance time , and verify the issuance time to ensure ; Step 212, record the execution results of each public opinion task, specifically including: public opinion task type, the corresponding public opinion keyword tags of this type, public opinion task issuance time, and issuance priority; Step 213, integrate and package the recorded information into a public opinion data packet for subsequent analysis.

[0021] The steps for the public opinion security situation awareness module to conduct situation analysis on the completed public opinion tasks, determine whether feedback events need to be formed based on the situation analysis results, and send the formed feedback events to the emergency response and maintenance module include: Step 301, obtain the public opinion data packet, and use the user behavior log to capture the response time, user participation, and public opinion task effect of each public opinion task in the public opinion data packet; In Step 301, the response time of the public opinion task represents the time delay of the public opinion release relative to the public opinion outbreak; the user participation represents the number of user interactions related to this public opinion task, such as the number of comments and likes; the public opinion task effect represents the evaluation of the social influence of this public opinion task, such as the spread range and media exposure; Step 302, extract the response time of the th public opinion task, the actual user participation of the th public opinion task, and the original effect value of the th public opinion task; Step 303, conduct situation analysis on the extracted public opinion task response time, actual user participation, and original effect value: Perform power-law scaling on the public opinion task response time to obtain , where represents the public opinion task response time after power-law scaling, denotes the power-law scaling exponent, which is a positive real number greater than 1 and is used to amplify the impact of the response time on the public opinion task. When there is a long-term anomaly in the response time of a certain public opinion task, a larger can significantly improve the value, emphasizing the negative effect brought by the long delay; Tracking the non-linear change trend of the actual user engagement to obtain , where represents the user engagement after non-linear change, represents the maximum possible user engagement level. Setting this upper limit helps to conduct standardized comparisons. represents the control of the steepness of the curve, which determines the tightness of the non-linear relationship between user engagement and public opinion tasks. A higher value indicates that if the user engagement is slightly lower, it will lead to a significant decrease in the steepness of the curve, and vice versa, it is more tolerant; Compressing the dynamic range of the original effect value of the public opinion task to obtain , where represents the effect value of the public opinion task after completing the dynamic range compression, is the arctangent function, represents the amplification coefficient, which is used to enhance the change amplitude of the original effect value of the public opinion task and achieve a more delicate resolution effect in a small range. It is especially suitable for highlighting those events with small effects but requiring attention; Step 304: Vertically compare the current situation analysis results with the historical public opinion data of similar topics. Determine whether a feedback event needs to be formed. If the current situation significantly deviates from the average level or trend line of the historical data, it indicates that there is an anomaly or a situation that needs attention, thus triggering the formation of a feedback event; In step 304, the historical public opinion data of similar topics refers to the public opinion of similar topics in other regions or industries.

[0022] The steps for the emergency response and maintenance module to conduct real-time monitoring and quantitative analysis of the feedback event and to process and control the parameters include: Step 401: Obtain the , and corresponding to the feedback event; Step 402: Real-time monitor the feedback event and assign a unique identifier to each monitored feedback event; Step 403: Use the formula to conduct quantitative analysis of the feedback event: where represents the comprehensive evaluation value of the feedback event; is the weight coefficient, which is used to adjust , and the proportion of the comprehensive evaluation of feedback events; is the sensitivity controller, which is used to control the influence speed. If has a larger value, it means that the influence is more intense; is the sensitivity controller, which is used to adjust the influence speed. If has a larger numerical value, it means that the influence is milder; is the hyperbolic tangent function, which is used to map to the interval from -1 to 1; Step 404, combine the comprehensive evaluation value of the feedback event to set the response threshold ; among them, the response threshold is used to distinguish feedback events of different emergency levels or importance; Step 405, when comparing the calculated comprehensive evaluation value of the feedback event with the preset response threshold : when the comprehensive evaluation value of the feedback event is higher than the response threshold , it means that the feedback event has a high emergency level, has a high degree of urgency or importance, may have a greater impact on society, organization or users, and it is necessary to immediately start the emergency response process and generate a warning prompt. The emergency response process specifically includes: sending a warning prompt to relevant personnel to ensure that they can quickly understand the public opinion situation; organizing forces to respond, including allocating resources and deploying personnel; collecting more information to further understand the development trend and potential impact of the public opinion; and adjusting the public opinion handling strategy according to the specific situation to ensure that the public opinion is properly controlled; when the comprehensive evaluation value of the feedback event is not higher than the response threshold , it means that the feedback event has a low emergency level and does not reach the level that requires an immediate response. Continue to monitor the dynamic changes of the feedback event, and at the same time formulate corresponding preventive measures for the feedback event to reduce potential risks; Step 406, after the current public opinion control ends, remove the public opinion keyword tags corresponding to the feedback event of the current identifier; In step 406, the feedback event of the current identifier represents the feedback event currently being processed.

[0023] In the embodiments of the present invention, by collecting the mainstream information of self-media on social media platforms and standardizing this information, the problem of inconsistent measurement scales among different self-media is solved, ensuring the accuracy and comparability of data. The communication power data is represented in multiple dimensions through a five-tuple (title, click-through rate, comment volume, like volume, repost volume), which can more comprehensively reflect the dissemination of public opinion information. By introducing a time weight function and a weight coefficient, considering the importance of different elements at different time points, a comprehensive communication power index is calculated, providing a scientific basis for public opinion analysis. By calculating a multi-dimensional influence index based on the comprehensive communication power index and combining the analysis of statistics such as mean and standard deviation, it is determined whether the multi-dimensional influence index is normal, effectively identifying potential public opinion outbreak points. By using TF-IDF analysis to screen high-frequency words and generating public opinion keyword tags, it helps to quickly grasp the public opinion hotspots. By automatically classifying public opinion task types (repost task, comment task, contribution task) according to the public opinion keyword tags, the efficiency and accuracy of public opinion processing are improved. By setting constraints for the issuance of public opinion tasks (minimum time interval, maximum time interval), it is avoided that the tasks are issued too densely or sparsely, ensuring the continuity and stability of public opinion processing. At the same time, the issuance order is determined according to the priority of public opinion tasks to ensure that important tasks are processed first. By conducting a situation analysis of the response time, user participation, and public opinion task effects of public opinion tasks, key information is highlighted, improving the accuracy of the analysis. By real-time monitoring and quantitatively analyzing feedback events, evaluating their urgency and importance, a scientific basis for emergency response is provided. By real-time monitoring feedback events, using formulas for quantitative analysis, and taking timely emergency treatment measures, it is ensured that public opinion is quickly responded to and properly controlled. In summary, the embodiments of the present invention solve the problem of inaccurate public opinion processing and analysis in current public opinion management. In actual situations, more data and context information may be required to make specific decisions and optimization plans.

[0024] In addition, the formulas involved above are all calculated by removing the dimension and taking their numerical values. They are obtained by collecting a large amount of data and performing software simulation to obtain a formula closest to the real situation. The proportionality coefficients in the formulas and various preset thresholds in the analysis process are set by those skilled in the art according to the actual situation or obtained through a large amount of data simulation; the size of the proportionality coefficient is a specific value obtained by quantifying each parameter for subsequent comparison. Regarding the size of the proportionality coefficient, it depends on the amount of sample data and the preliminary setting of the corresponding processing coefficients for each group of sample data by those skilled in the art; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0025] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the device embodiments, since they are basically based on the method embodiments, they are described relatively simply, and reference can be made to the corresponding parts of the method embodiments for the relevant content.

[0026] For the convenience of description, when describing the above device, various units are described separately according to their functions. Of course, when implementing the present application, the functions of each unit can be realized in one or more software and / or hardware.

[0027] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of an all-hardware embodiment, an all-software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0028] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or a combination of multiple flows and / or blocks.

[0029] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or a combination of multiple flows and / or blocks.

[0030] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide for implementing the functions in Figure 1Steps of the functions specified in one process or multiple processes and / or boxes Figure 1 Steps of the functions specified in one box or multiple boxes

[0031] Secondly: In the accompanying drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments of the present disclosure are involved. For other structures, reference may be made to the general design. Without conflict, the same embodiment and different embodiments of the present invention may be combined with each other; Finally: The above is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent replacements or changes, and should be covered by the protection scope of the present invention.

Claims

1. An information management platform based on an opinion handling process, characterized in that: It includes a communication power data analysis module, a public opinion processing interaction module, a public opinion security situation awareness module, and an emergency response and maintenance module; Communication power data analysis module: Obtain communication power data, perform statistical analysis on it to obtain a comprehensive communication power index, calculate a multi-dimensional influence index based on the comprehensive communication power index, judge whether the multi-dimensional influence index is normal, and generate public opinion keyword tags; Public opinion processing interaction module: Obtain public opinion keyword tags, automatically divide public opinion task types for the public opinion keyword tags, set and issue corresponding public opinion tasks based on the divided public opinion task types, and determine the time interval for the execution of public opinion tasks; among them, the public opinion processing interaction module includes a public opinion task division unit and a public opinion task execution unit; Public opinion security situation awareness module: Conduct a situation analysis on the completed public opinion tasks, judge whether a feedback event needs to be formed according to the situation analysis result, and send the formed feedback event to the emergency response and maintenance module; Emergency response and maintenance module: Conduct real-time monitoring and quantitative analysis on the feedback event, and perform processing and control.

2. The information management platform based on the public opinion processing process according to claim 1, characterized in that: The process by which the communication power data analysis module obtains communication power data and performs statistical analysis on it to obtain a comprehensive communication power index includes: Collect mainstream self-media information from social media platforms and standardize the information published by self-media; among them, the mainstream self-media information includes title, click-through rate, comment volume, like volume, and repost volume; Mark the mainstream self-media information after standardization as communication power data , and use a five-tuple to represent the marked communication power data for multi-dimensional representation: ; among them, the five-tuple consists of five elements: title , click-through rate , comment volume , like volume , and repost volume ; Statistically analyze all communication power data in chronological order and integrate to obtain a comprehensive communication power index: In the formula, represents the comprehensive dissemination power index; represents each element of the dissemination power data, that is ; represents the quantity of each element of the dissemination power data; represents the weight coefficient of each element in the dissemination power data; represents the time weight function of each element in the dissemination power data.

3. An information management platform based on the public opinion processing process according to claim 1, characterized in that: The process by which the communication power data analysis module calculates a multi-dimensional influence index based on the comprehensive communication power index, judges whether the multi-dimensional influence index is normal, and generates public opinion keyword tags includes: Calculate and analyze the multi-dimensional influence index according to the comprehensive communication power index: Formula represents a multi-dimensional influence index; is a preset proportionality coefficient; represents the time window size for calculating the average value; is an index variable; is a non-linear effect function, evaluated based on the second derivative of, expressed as: , where are respectively the first derivative and the second derivative of, representing the rate of change and the acceleration of; When analyzing the calculation results of the multi-dimensional influence index all multi-dimensional influence indexes are sorted in descending order according to the numerical values of the multi-dimensional influence index to obtain the mean value and the standard deviation ; Combined with the mean of the multi-dimensional influence index and the standard deviation Judge and analyze whether the multi-dimensional influence index is within the normal range; Among them, , , where in the formula, represents the number of multi-dimensional influence indices, represents the index of the multi-dimensional influence index; set the detection limit value of the outlier, and consider the multi-dimensional influence index higher than as an outlier, otherwise, consider it as a normal value. Where in the formula, represents the selected multiple used to define the outlier; Collect all the multi-dimensional influence indices of anomalies and determine the comprehensive dissemination power indicators corresponding to the anomalous multi-dimensional influence indices ; Collect this comprehensive communication power index Matched text data, including news titles, comments, or social media posts, and organize this text data into public opinion information; Use a text analysis tool to perform TF-IDF analysis on the associated text and screen out high-frequency words: In the formula, represents a term; represents a document; represents the term appearing in the document the number of times; represents the document the total number of terms; represents the term appearing in the document the number of times and the total number of terms of this document i.e., the term frequency; represents the entire document collection the total number; represents the number of documents containing the term ; represents the inverse document frequency; Identify words that are high-frequency and have high discrimination for the document set, and select such words as public opinion keywords; Generate corresponding public opinion keyword tags in combination with the public opinion keywords.

4. An information management platform based on an opinion handling process according to claim 1, characterized in that: The process by which the public opinion task division unit automatically divides public opinion task types for the public opinion keyword tags includes: Obtain public opinion keyword tags and set a public opinion keyword tag set ; According to the public opinion information containing the public opinion keyword tag set , automatically divide the public opinion task types. Each public opinion task type includes a forwarding task, a commenting task, and a contribution task, and only one public opinion task type is allowed to be selected each time a public opinion task is issued.

5. An information management platform based on an opinion handling process according to claim 1, characterized in that: The process by which the public opinion task execution unit sets and issues corresponding public opinion tasks and determines the time interval for the execution of public opinion tasks based on the divided public opinion task types includes: Suppose there is a set of public opinion tasks that need to be assigned; among them, represents the total number of public opinion tasks Each public opinion task has a corresponding fixed distribution time ; among them, represents the public opinion task index; Define the time interval between the current public opinion task and the next public opinion task as ; Set the constraint conditions for the distribution of public opinion tasks: the minimum time interval and the maximum time interval , which is used to limit the frequency of the distribution of public opinion tasks; Assign a corresponding distribution priority to each public opinion task , and determine the order of distribution of public opinion tasks according to the priorities of public opinion tasks; Initialize the distribution time for the first public opinion task Set the corresponding initial distribution time ; For each subsequent public opinion task , calculate the time interval between the current public opinion task and the next public opinion task ; Minimum time interval according to the constraint condition and maximum time interval Adjust the time interval ; Calculate the next public opinion task for the release time ; Record the execution results of each public opinion task, specifically including: public opinion task type, the corresponding public opinion keyword tag of this type, public opinion task issuance time, and issuance priority; Integrate and package the recorded information into a public opinion data packet.

6. The information management platform based on the public opinion processing process according to claim 1, wherein: The process by which the public opinion security situation awareness module conducts a situation analysis on the completed public opinion tasks and judges whether a feedback event needs to be formed according to the situation analysis result includes: Obtain the public opinion data packet, use the user behavior log to capture the response time, user participation, and public opinion task effect of each public opinion task in the public opinion data packet; among them, the response time of the public opinion task represents the time delay of the public opinion release relative to the public opinion outbreak; user participation represents the number of user interactions related to this public opinion task; the public opinion task effect represents the evaluation of the social influence of this public opinion task; Extract the response time of the first public opinion task, the actual user participation of the second public opinion task, and the original effect value of the third public opinion task ; Perform situation analysis on the extracted public opinion task response time, actual user participation, and original effect value: The response time of the public opinion task is power-law scaled to obtain , where represents the response time of the public opinion task after power-law scaling, represents the power-law scaling exponent, which is a positive real number greater than 1 and is used to amplify the impact of the response time on the public opinion task; Tracking the non-linear change trend of actual user engagement to obtain , where represents the user engagement after non-linear change, represents the maximum possible user engagement level, represents the control of the steepness of the curve, which determines the tightness of the non-linear relationship between user engagement and public opinion tasks; The dynamic range of the original effect value of the public opinion task is compressed to obtain , where represents the effect value of the public opinion task after completing dynamic range compression, is the arctangent function, represents the amplification factor, which is used to enhance the change range of the original effect value of the public opinion task; The current situation analysis results are compared vertically with the historical public opinion data of similar topics to determine whether a feedback event needs to be formed. If the current situation deviates significantly from the average level or trend line of historical data, it indicates that there is an abnormality or a situation that requires attention, thereby triggering the formation of a feedback event.

7. An information management platform based on the public opinion processing process according to claim 1 or 6, characterized in that: The emergency response and maintenance module conducts real-time monitoring and quantitative analysis of feedback events, and the process of processing and control includes: Obtain those corresponding to the feedback event , and ; Monitor feedback events in real time and assign a unique identifier to each monitored feedback event; Use the formula to quantify feedback events: In the formula, represents the comprehensive evaluation value of the feedback event; are weight coefficients, respectively used to adjust , and the proportion of the comprehensive evaluation of the feedback event; is a sensitivity controller, used to control the influence speed; is a sensitivity controller, used to adjust the influence speed; is the hyperbolic tangent function, used to map to the interval from -1 to 1; Comprehensive evaluation value combined with feedback events Set the response threshold ; Among them, the response threshold is used to distinguish feedback events of different levels of urgency or importance; Based on the calculated comprehensive evaluation value of the feedback event and the preset response threshold when comparing: when the comprehensive evaluation value of the feedback event is higher than the response threshold it indicates that the feedback event has a high urgency level, and it is necessary to immediately initiate the emergency response process and generate a warning prompt; when the comprehensive evaluation value of the feedback event is not higher than the response threshold it indicates that the feedback event has a low urgency level, continue to monitor the dynamic changes of the feedback event, and at the same time formulate corresponding preventive measures for the feedback event; After this public opinion control is completed, the public opinion keyword label corresponding to the current identifier feedback event is removed; among which, the current identifier feedback event indicates the feedback event currently being processed.

Images