Consumption public opinion monitoring system

Abstract

The invention discloses a consumption public opinion monitoring system, which automatically captures wide consumption data on the Internet through a web crawler technology, is safely connected with an enterprise internal database to extract complaint records, ensures that the data is comprehensive, accurate and safe, and automatically classifies the data according to a preset classification system after cleaning processing and redundancy and error removal. The processing efficiency is improved; the emotion analysis algorithm is used for judging the emotion tendency of the complaint text and understanding the customer satisfaction; the topic model algorithm is used for mining main complaint topics and keywords, quickly mastering the consumption public opinions, analyzing and revealing the complaint hotspot trend through a time sequence, providing prospective decision support, and combining a geographic information system and customer portraits; the method analyzes differences between regions and consumer groups, formulates accurate policies, provides various report templates, meets diversified demands of users, automatically generates and pushes public opinion monitoring reports, ensures timely and effective information, assists enterprises and the government in timely response to consumption public opinions, and improves public satisfaction and credibility.

Description

Technical Field

[0001] The present invention relates to the technical field of public opinion monitoring, and in particular to a consumer public opinion monitoring system. Background Art

[0002] At present, in the field of consumer public opinion monitoring, some commonly used monitoring technologies or software have many shortcomings. Some traditional public opinion monitoring tools mainly focus on the collection of a single data source. For example, they only capture data from social media platforms or specific news websites. They are unable to fully integrate the scattered complaint public opinions on the Internet with the consumer complaint data of enterprises or platforms themselves. As a result, the data sources are relatively limited, making it difficult to form a complete and accurate portrait of consumer public opinion.

[0003] Some existing analysis software has relatively simple functions. When analyzing the collected data, it can often only perform simple keyword statistics or basic sentiment tendency judgments, and lacks in-depth data mining capabilities. For example, it is difficult to accurately identify the evolution trend of complaint hotspots, differences in complaint characteristics between different regions and consumer groups, and other complex analysis functions. It is unable to provide comprehensive and forward-looking decision-making basis for enterprises or regulatory authorities.

[0004] In addition, in terms of report generation, most existing technologies generate reports with fixed formats and brief content, and only present some basic data statistical results. They lack analysis of the deep-seated reasons behind the data and targeted recommended measures, making it difficult to meet users' needs for detailed, professional and actionable reports. Moreover, these existing technologies also have defects in the timeliness of data updates, and are unable to track the dynamic changes of public opinion in real time, resulting in delayed response when dealing with sudden consumer public opinion events. Summary of the invention

[0005] In view of this, the present invention proposes a consumer public opinion monitoring system, which can effectively solve the many problems faced by the existing technology in consumer public opinion monitoring, improve the monitoring and management efficiency of the consumer market, protect the rights and interests of consumers and help enterprises optimize products and services.

[0006] The technical solution of the present invention is achieved in this way:

[0007] A consumer public opinion monitoring system, comprising:

[0008] The data collection module is used to regularly capture consumption-related data using web crawler technology and directly extract consumer complaint records by establishing a secure connection with the database within the enterprise or platform;

[0009] The data preprocessing module is used to clean the collected data, automatically classify the cleaned data according to the pre-set consumer complaint classification system, and manually mark the key data;

[0010] The text analysis module is used to use the sentiment analysis algorithm in natural language processing technology to judge the sentiment tendency of complaint texts, and adopt the topic model algorithm to automatically mine the main complaint topics and related keywords from a large number of complaint texts;

[0011] The data mining module is used to collect and analyze complaint data in different time periods based on time series analysis methods, predict the development trend of complaint hotspots, and analyze the differences in complaint characteristics in different regions and consumer groups by combining geographic information system technology and consumer portrait data;

[0012] The data reporting module is used to provide a variety of report templates for corporate and government users to choose from, and automatically generate a complete consumer public opinion monitoring report based on the selected templates and visual data, and send it in a timely manner according to the push method set by the user.

[0013] As a further optional solution of the consumer public opinion monitoring system, the data collection module uses web crawler technology to regularly capture consumption-related data and directly extracts consumer complaint records by establishing a secure connection with the database within the enterprise or platform, specifically including:

[0014] Use a distributed web crawler architecture to target major social platforms, consumer complaint forums, and e-commerce platform comment areas, set specific keywords, time ranges, and geographical ranges for screening, and regularly capture consumer-related complaint texts, images, and video data;

[0015] By establishing a secure connection with the database within the enterprise or platform, consumer complaint records can be directly extracted. The consumer complaint records include complaint time, complaint content, complaining user information, and involved product and service categories.

[0016] As a further optional solution of the consumer public opinion monitoring system, the data preprocessing module performs data cleaning on the collected data, automatically classifies the cleaned data according to a pre-set consumer complaint classification system, and manually labels key data, specifically including:

[0017] A consumer complaint classification system with pre-classification algorithms based on common consumer complaint issues and areas;

[0018] Unify the format of all types of collected data and remove noise information in the data;

[0019] During the data cleaning process, duplicate data records are identified and removed;

[0020] Extract features from the cleaned data, where the features include keywords, phrases, sentence structures, and sentiment tendencies in the text;

[0021] The extracted features are input into the classification algorithm, and the classification algorithm automatically classifies according to the preset classification system and features;

[0022] Manually label key data.

[0023] As a further optional solution of the consumer public opinion monitoring system, the text analysis module uses the sentiment analysis algorithm in natural language processing technology to judge the sentiment tendency of the complaint text, and adopts the topic model algorithm to automatically mine the main complaint topics and related keywords from a large number of complaint texts, including:

[0024] Using a trained sentiment analysis model, wherein the sentiment analysis model is trained based on consumer complaint text data with clear sentiment labels;

[0025] The preprocessed complaint text is input into the sentiment analysis model, and the sentiment analysis model outputs the corresponding sentiment category and sentiment intensity score;

[0026] The preprocessed complaint text is input into the topic model algorithm, which automatically mines the topic structure of the text based on the probability statistics principle of the co-occurrence relationship of words in the text;

[0027] Output the keywords corresponding to each topic and the distribution probability of each topic in the text data.

[0028] As a further optional solution of the consumer public opinion monitoring system, the data mining module is based on the time series analysis method to collect and analyze the complaint data in different time periods, predict the development trend of complaint hotspots, and combine geographic information system technology and consumer portrait data to analyze the differences in complaint characteristics of different regions and different consumer groups, including:

[0029] Select the appropriate time series model based on the characteristics of complaint data in different time periods;

[0030] Use time series models to fit and predict complaint data and predict the development trend of complaint hotspots within a preset time period;

[0031] Collect geographic data related to complaint data;

[0032] Associating complaint data with geographic data to form a complaint dataset containing geographic information;

[0033] Use GIS technology to visualize the complaint dataset;

[0034] Collect consumer portrait data;

[0035] The consumer portrait data is associated with the complaint data set to analyze the differences in complaint characteristics of different consumer groups in different regions.

[0036] As a further optional solution of the consumer public opinion monitoring system, the data reporting module provides a variety of report templates for enterprise and government users to choose from, and automatically generates a complete consumer public opinion monitoring report based on the selected template and visualized data, and sends it in a timely manner according to the push method set by the user, specifically including:

[0037] Use online public opinion monitoring tools to monitor public opinion information related to consumer sentiment in real time;

[0038] Screen and filter the collected public opinion information, extract and organize the information related to the monitored object, and classify and label it;

[0039] Conduct in-depth analysis of the filtered information, analyze the hot issues of public opinion, as well as the development trends and changing patterns of public opinion;

[0040] Automatically generate consumer public opinion monitoring reports based on the selected template and set parameters;

[0041] According to the push method and time set by the user, the public opinion monitoring report is automatically generated and sent to the receiving end specified by the user.

[0042] As a further optional solution for the consumer public opinion monitoring system, the system also includes a system management and maintenance module, which is used to set different authority levels for enterprise and government users, regularly and automatically update data collection sources, perform regular data backup, and monitor the system's operating status in real time.

[0043] A method for monitoring consumer public opinion, specifically comprising:

[0044] Use web crawler technology to regularly crawl consumption-related data from the Internet. At the same time, establish a secure connection with the database within the enterprise or platform to directly extract consumer complaint records;

[0045] Clean the collected data, automatically classify the cleaned data according to the pre-set consumer complaint classification system, and manually mark the key data;

[0046] Use sentiment analysis algorithms in natural language processing technology to judge the sentiment tendency of complaint texts and distinguish positive, negative or neutral comments. At the same time, use topic model algorithms to automatically mine the main complaint topics and related keywords from a large number of complaint texts;

[0047] Based on the time series analysis method, the complaint data in different time periods are counted and analyzed to reveal the development trend and change law of complaint hotspots. Combined with geographic information system technology, the differences in complaint characteristics in different regions are analyzed. At the same time, combined with consumer portrait data, the differences in complaint characteristics of different consumer groups are analyzed.

[0048] Based on the selected template and visualized data, a complete consumer opinion and public opinion monitoring report is automatically generated, and the consumer opinion and public opinion monitoring report is sent to relevant users in a timely manner according to the push method set by the user.

[0049] A computing device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the steps of the above-mentioned method for monitoring consumer public opinion when executing the computer program.

[0050] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any one of the above-mentioned methods for monitoring consumer public opinion.

[0051] The beneficial effects of the present invention are as follows: through web crawler technology, the system can regularly and automatically capture a wide range of consumption-related data from the Internet, including consumer evaluations, news reports, social media discussions, etc., ensuring the comprehensiveness and timeliness of the data, establishing a secure connection with the database within the enterprise or platform, and directly extracting consumer complaint records, which not only improves the accuracy of the data, but also avoids the risk of data leakage and enhances data security. The data cleaning process effectively removes duplicate, invalid or erroneous data and improves data quality. The cleaned data is automatically classified according to a pre-set consumer complaint classification system, which greatly saves the time and cost of manual classification and improves the efficiency of data processing. The key data is manually labeled to ensure the accuracy and reliability of the data. The sentiment analysis algorithm in natural language processing technology is used to judge the sentiment tendency of the complaint text, which can accurately identify the positive, negative or neutral evaluation of consumers, which helps to understand the satisfaction and emotional changes of consumers. The topic model algorithm is used to automatically classify the complaint text from a large number of complaint texts. Automatically digging out the main complaint topics and related keywords can help companies and governments quickly understand the main concerns of consumer public opinion and take timely measures to solve problems. Based on the time series analysis method, the complaint data in different time periods are counted and analyzed, which can reveal the development trend and change law of complaint hotspots and provide forward-looking decision-making support for companies and governments. Combining geographic information system technology and consumer portrait data, the differences in complaint characteristics of different regions and consumer groups are analyzed, which is helpful to formulate more accurate regional consumer policies and marketing strategies. A variety of report templates are provided for corporate and government users to choose from, which meets the needs of different users for report format and content, and improves the practicality and readability of the report. According to the selected template and visualized data, a complete consumer public opinion monitoring report is automatically generated, which greatly saves the time and cost of manual report writing. The report is sent in time according to the push method set by the user, ensuring the timeliness and effectiveness of the information, which helps companies and governments respond to consumer public opinion in a timely manner and improve public satisfaction and trust. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 A schematic diagram of the composition of a consumer public opinion monitoring system of the present invention;

[0054] Figure 2 A schematic diagram of a process of monitoring consumer public opinion and sentiment according to the present invention;

[0055] Figure 3 The figure is a schematic diagram of the composition of a computing device according to the present invention. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] refer to Figures 1 to 3 , a consumer public opinion monitoring system, comprising:

[0058] The data collection module is used to regularly capture consumption-related data using web crawler technology and directly extract consumer complaint records by establishing a secure connection with the database within the enterprise or platform;

[0059] The data preprocessing module is used to clean the collected data, automatically classify the cleaned data according to the pre-set consumer complaint classification system, and manually mark the key data;

[0060] The text analysis module is used to use the sentiment analysis algorithm in natural language processing technology to judge the sentiment tendency of complaint texts, and adopt the topic model algorithm to automatically mine the main complaint topics and related keywords from a large number of complaint texts;

[0061] The data mining module is used to collect and analyze complaint data in different time periods based on time series analysis methods, predict the development trend of complaint hotspots, and analyze the differences in complaint characteristics in different regions and consumer groups by combining geographic information system technology and consumer portrait data;

[0062] The data reporting module is used to provide a variety of report templates for corporate and government users to choose from, and automatically generate a complete consumer public opinion monitoring report based on the selected templates and visual data, and send it in a timely manner according to the push method set by the user.

[0063] In this embodiment, through web crawler technology, the system can regularly and automatically capture a wide range of consumption-related data from the Internet, including consumer evaluations, news reports, social media discussions, etc., ensuring the comprehensiveness and timeliness of the data, establishing a secure connection with the database within the enterprise or platform, and directly extracting consumer complaint records, which not only improves the accuracy of the data, but also avoids the risk of data leakage and enhances data security. The data cleaning process effectively removes duplicate, invalid or erroneous data and improves data quality. The cleaned data is automatically classified according to a pre-set consumer complaint classification system, which greatly saves the time and cost of manual classification and improves the efficiency of data processing. Key data is manually labeled to ensure the accuracy and reliability of the data. The sentiment analysis algorithm in natural language processing technology is used to judge the sentiment tendency of the complaint text, which can accurately identify consumers' positive, negative or neutral evaluations, which helps to understand consumers' satisfaction and emotional changes. The topic model algorithm is used to automatically mine from a large amount of complaint text. Unearthing the main complaint topics and related keywords can help companies and governments quickly understand the main concerns of consumer public opinion and take timely measures to solve problems. Based on the time series analysis method, statistics and analysis of complaint data in different time periods can reveal the development trend and change pattern of complaint hotspots, and provide forward-looking decision-making support for companies and governments. Combining geographic information system technology and consumer portrait data, analyzing the differences in complaint characteristics in different regions and consumer groups will help to formulate more accurate regional consumer policies and marketing strategies. A variety of report templates are provided for corporate and government users to choose from, meeting the needs of different users for report format and content, and improving the practicality and readability of the report. According to the selected template and visualized data, a complete consumer public opinion monitoring report is automatically generated, which greatly saves the time and cost of manual report writing. The report is sent in a timely manner according to the push method set by the user, ensuring the timeliness and effectiveness of the information, which helps companies and governments respond to consumer public opinion in a timely manner and improve public satisfaction and trust.

[0064] Preferably, the data collection module uses web crawler technology to regularly capture consumption-related data, and directly extracts consumption complaint records by establishing a secure connection with the database within the enterprise or platform, specifically including:

[0065] Use a distributed web crawler architecture to target major social platforms, consumer complaint forums, and e-commerce platform comment areas, set specific keywords, time ranges, and geographical ranges for screening, and regularly capture consumer-related complaint texts, images, and video data;

[0066] By establishing a secure connection with the database within the enterprise or platform, consumer complaint records can be directly extracted. The consumer complaint records include complaint time, complaint content, complaining user information, and involved product and service categories.

[0067] In this embodiment, a distributed architecture is adopted, which can capture data in parallel from multiple nodes at the same time, significantly improving the speed and efficiency of data collection. The distributed architecture also has good scalability and fault tolerance, and can adapt to data collection tasks of different scales and complexities; setting specific keywords, time ranges, and geographical ranges for screening conditions can accurately locate the target data, avoid the collection of invalid data, and regularly capture data to ensure the timeliness and freshness of the data, which helps to timely discover and analyze changes in consumer public opinion; data collection is carried out on major mainstream social platforms and consumer complaint forums, which can capture the real voice and feedback of consumers on these platforms. These platforms usually contain a large amount of consumer evaluations, complaints, discussions and other information, and are important windows for understanding consumer public opinion; e-commerce platforms The comment area is an important place for consumers to post product reviews and usage experiences. By capturing this data, we can gain an in-depth understanding of consumers' satisfaction and opinions on products and services. In addition to text data, multimedia data such as pictures and videos are also captured. These data can often more intuitively display consumers' usage experience and complaints. By establishing a secure connection with the database within the enterprise or platform, consumer complaint records can be directly extracted, including key information such as complaint time, complaint content, complaining user information, and product and service categories involved. This method avoids data loss and deformation during transmission and ensures data accuracy and integrity. When connecting to and extracting data from the internal database of the enterprise or platform, strictly abide by data security specifications and privacy protection regulations to ensure the security and privacy of user data.

[0068] Preferably, the data preprocessing module performs data cleaning on the collected data, automatically classifies the cleaned data according to a pre-set consumer complaint classification system, and manually labels key data, specifically including:

[0069] A consumer complaint classification system with pre-classification algorithms based on common consumer complaint issues and areas;

[0070] Unify the format of all types of collected data and remove noise information in the data;

[0071] During the data cleaning process, duplicate data records are identified and removed;

[0072] Extract features from the cleaned data, where the features include keywords, phrases, sentence structures, and sentiment tendencies in the text;

[0073] The extracted features are input into the classification algorithm, and the classification algorithm automatically classifies according to the preset classification system and features;

[0074] Manually label key data.

[0075] In this embodiment, the various types of collected data are processed in a unified format to ensure the consistency and comparability of the data. The unified format processing also includes the conversion of data types, the processing of missing values, etc. These steps help to improve the quality of data and the accuracy of analysis; noise information usually refers to useless, misleading or redundant parts of the data. By removing this data, the interference of data analysis can be reduced and the accuracy of analysis can be improved. The existence of duplicate data will lead to deviations in data analysis results. By identifying and removing duplicate data, the uniqueness of the data set and the accuracy of the analysis are ensured; based on the common problems and fields of consumer complaints, a consumer complaint classification system with classification algorithms is pre-established. This system provides clear guidance for the automatic classification of data, ensuring the accuracy and consistency of classification. Feature extraction is performed on the cleaned data, including keywords, phrases, sentence structures and emotional tendencies in the text. These features can reflect the essence and core information of the data. The extracted features are input into the classification algorithm, and the classification algorithm automatically classifies according to the preset classification system and features. This process realizes the intelligent processing of data and improves the efficiency and accuracy of classification. Although the automatic classification algorithm can process most of the data, some complex or ambiguous data still needs manual labeling and confirmation. This step ensures the accuracy and reliability of data processing. Through manual labeling, errors in automatic classification can be corrected, and the quality of data and the accuracy of analysis can be improved. At the same time, manual labeling can also provide more valuable reference information for subsequent text analysis and data mining.

[0076] Preferably, the text analysis module uses the sentiment analysis algorithm in natural language processing technology to judge the sentiment tendency of the complaint text, and adopts the topic model algorithm to automatically mine the main complaint topics and related keywords from a large number of complaint texts, specifically including:

[0077] Using a trained sentiment analysis model, wherein the sentiment analysis model is trained based on consumer complaint text data with clear sentiment labels;

[0078] The preprocessed complaint text is input into the sentiment analysis model, and the sentiment analysis model outputs the corresponding sentiment category and sentiment intensity score;

[0079] The preprocessed complaint text is input into the topic model algorithm, which automatically mines the topic structure of the text based on the probability statistics principle of the co-occurrence relationship of words in the text;

[0080] Output the keywords corresponding to each topic and the distribution probability of each topic in the text data.

[0081] In this embodiment, by using a trained sentiment analysis model, the model is trained based on consumer complaint text data with clear sentiment labels, so that the sentiment tendency (such as positive, negative or neutral) in the complaint text can be accurately identified, and the output of the sentiment intensity score further refines the judgment of the sentiment tendency, making the analysis result more specific and quantitative; the topic model algorithm is based on the probabilistic statistical principle of word co-occurrence relationship, and can automatically mine the main complaint topics from a large number of complaint texts. This automated processing method greatly improves the processing efficiency and reduces the time and cost of manual analysis; the keyword extraction corresponding to each topic makes the main content of the complaint clear at a glance, which helps to quickly understand the core issues of the complaint. The extraction of keywords also provides key information support for subsequent analysis and decision-making; the distribution probability of each topic in the text data reveals the prevalence and importance of different complaint topics. This comprehensive analysis helps to identify the most common or most urgent complaint issues, thereby guiding enterprises to take targeted improvement measures; by deeply understanding customers' sentiment tendencies and complaint topics, enterprises can more accurately locate customer needs and dissatisfaction points, which helps enterprises optimize customer service processes, improve customer satisfaction, and at the same time improve products or services and reduce the occurrence of complaints.

[0082] Preferably, the data mining module collects and analyzes complaint data in different time periods based on a time series analysis method, predicts the development trend of complaint hotspots, and analyzes the differences in complaint characteristics of different regions and consumer groups by combining geographic information system technology and consumer portrait data, specifically including:

[0083] Select the appropriate time series model based on the characteristics of complaint data in different time periods;

[0084] Use time series models to fit and predict complaint data and predict the development trend of complaint hotspots within a preset time period;

[0085] Collect geographic data related to complaint data;

[0086] Associating complaint data with geographic data to form a complaint dataset containing geographic information;

[0087] Use GIS technology to visualize the complaint dataset;

[0088] Collect consumer portrait data;

[0089] The consumer portrait data is associated with the complaint data set to analyze the differences in complaint characteristics of different consumer groups in different regions.

[0090] In this embodiment, according to the characteristics of complaint data in different time periods, a suitable time series model is selected for fitting and prediction, which can accurately capture the changing trend of complaint hotspots. This prediction capability helps enterprises identify potential complaint risks in advance and provide a time window for formulating response strategies; by associating complaint data with geographic data and using GIS technology for visualization, the distribution of complaints in different regions can be intuitively displayed. This visualization analysis helps enterprises quickly identify complaint hotspots and provide guidance for resource allocation and on-site processing; combined with consumer portrait data, in-depth analysis can be made of the differences in complaint characteristics of different consumer groups in different regions. This deep insight helps enterprises understand consumer needs more accurately and provide a basis for personalized services and product improvements; through automated and intelligent data analysis processes, this technical solution significantly improves the processing efficiency of complaint data. This efficiency improvement helps enterprises quickly respond to consumer complaints, reduce complaint processing time, and improve overall operational efficiency.

[0091] Preferably, the data reporting module provides a variety of report templates for enterprise and government users to choose from, and automatically generates a complete consumer public opinion monitoring report based on the selected template and visualized data, and sends it in a timely manner according to the push method set by the user, specifically including:

[0092] Use online public opinion monitoring tools to monitor public opinion information related to consumer sentiment in real time;

[0093] Screen and filter the collected public opinion information, extract and organize the information related to the monitored object, and classify and label it;

[0094] Conduct in-depth analysis of the filtered information, analyze the hot issues of public opinion, as well as the development trends and changing patterns of public opinion;

[0095] Automatically generate consumer public opinion monitoring reports based on the selected template and set parameters;

[0096] According to the push method and time set by the user, the public opinion monitoring report is automatically generated and sent to the receiving end specified by the user.

[0097] In this embodiment, through the network public opinion monitoring tool, public opinion information related to consumer public opinion can be monitored in real time to ensure that enterprises and governments can obtain market dynamics and consumer feedback in the first time. This real-time monitoring capability helps to quickly respond to market changes and consumer needs and improve decision-making efficiency and accuracy; the collected public opinion information is screened and filtered, and information related to the monitored object is effectively extracted, and classified and labeled, which helps users to quickly understand the core content and importance of public opinion information. This information processing capability reduces the burden of users processing a large amount of irrelevant information and improves work efficiency; in-depth analysis of the screened information can reveal hot issues, development trends and changing laws of public opinion, and provide users with valuable insights and predictions. This analysis capability helps users to formulate more accurate and effective market strategies; according to The selected template and set parameters can automatically generate consumer public opinion monitoring reports, greatly reducing the time and cost of manual report writing. At the same time, it provides a variety of report templates for users to choose from, supports personalized customization, and meets the needs and preferences of different users; according to the push method and time set by the user, it automatically generates and sends the public opinion monitoring report to the user's designated receiving end to ensure that the user can obtain the report content in a timely manner. This push mechanism improves the report delivery efficiency and the user's receiving experience, and helps users make decisions and adjustments in a timely manner; the generated consumer public opinion monitoring report provides companies and governments with comprehensive market information and consumer feedback, which helps to make more scientific and reasonable decisions. At the same time, by monitoring changes in public opinion information, it can timely discover potential market risks and consumer dissatisfaction, and provide early warning and basis for risk management.

[0098] Preferably, the system also includes a system management and maintenance module for setting different permission levels for enterprise-side and government-side users, regularly and automatically updating data collection sources, performing scheduled data backup, and monitoring the operating status of the system in real time.

[0099] In this embodiment, by setting different permission levels for enterprise and government users, it is possible to ensure that information is accessed and used within an appropriate scope to prevent information leakage and abuse. This permission management mechanism helps to improve the security and controllability of the system; regularly and automatically update data collection sources to ensure that the system can continuously obtain the latest public opinion information and market trends. This update mechanism helps to maintain the timeliness and accuracy of the data; regular data backup can prevent data loss due to system failures, human errors or malicious attacks. In the event of data loss or damage, it can be restored through backup data to ensure the normal operation of the system and the integrity of the data; real-time monitoring of the system's operating status can promptly discover and resolve potential system problems and failures. This monitoring mechanism helps to improve the stability and reliability of the system and reduce the risks of service interruptions and data loss due to system failures; through automated data collection source updates, data backup and system operating status monitoring functions, manual intervention and maintenance costs can be reduced.

[0100] A method for monitoring consumer public opinion, specifically comprising:

[0101] Use web crawler technology to regularly crawl consumption-related data from the Internet. At the same time, establish a secure connection with the database within the enterprise or platform to directly extract consumer complaint records;

[0102] Clean the collected data, automatically classify the cleaned data according to the pre-set consumer complaint classification system, and manually mark the key data;

[0103] Use sentiment analysis algorithms in natural language processing technology to judge the sentiment tendency of complaint texts and distinguish positive, negative or neutral comments. At the same time, use topic model algorithms to automatically mine the main complaint topics and related keywords from a large number of complaint texts;

[0104] Based on the time series analysis method, the complaint data in different time periods are counted and analyzed to reveal the development trend and change law of complaint hotspots. Combined with geographic information system technology, the differences in complaint characteristics in different regions are analyzed. At the same time, combined with consumer portrait data, the differences in complaint characteristics of different consumer groups are analyzed.

[0105] Based on the selected template and visualized data, a complete consumer opinion and public opinion monitoring report is automatically generated, and the consumer opinion and public opinion monitoring report is sent to relevant users in a timely manner according to the push method set by the user.

[0106] A computing device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the steps of the above-mentioned method for monitoring consumer public opinion when executing the computer program.

[0107] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any one of the above-mentioned methods for monitoring consumer public opinion.

[0108] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A consumer public opinion monitoring system, characterized in that: include: The data collection module is used to regularly capture consumption-related data using web crawler technology and directly extract consumer complaint records by establishing a secure connection with the database within the enterprise or platform; The data preprocessing module is used to clean the collected data, automatically classify the cleaned data according to the pre-set consumer complaint classification system, and manually mark the key data; The text analysis module is used to use the sentiment analysis algorithm in natural language processing technology to judge the sentiment tendency of complaint texts, and adopt the topic model algorithm to automatically mine the main complaint topics and related keywords from a large number of complaint texts; The data mining module is used to collect and analyze complaint data in different time periods based on time series analysis methods, predict the development trend of complaint hotspots, and analyze the differences in complaint characteristics in different regions and consumer groups by combining geographic information system technology and consumer portrait data; The data reporting module is used to provide a variety of report templates for corporate and government users to choose from, and automatically generate a complete consumer public opinion monitoring report based on the selected templates and visual data, and send it in a timely manner according to the push method set by the user.

2. According to claim 1, a consumer public opinion monitoring system is characterized in that: The data collection module uses web crawler technology to regularly capture consumption-related data and directly extracts consumer complaint records by establishing a secure connection with the database within the enterprise or platform, including: Use a distributed web crawler architecture to target major social platforms, consumer complaint forums, and e-commerce platform comment areas, set specific keywords, time ranges, and geographical ranges for screening, and regularly capture consumer-related complaint texts, images, and video data; By establishing a secure connection with the database within the enterprise or platform, consumer complaint records can be directly extracted. The consumer complaint records include complaint time, complaint content, complaining user information, and involved product and service categories.

3. A consumer public opinion monitoring system according to claim 2, characterized in that: The data preprocessing module performs data cleaning on the collected data, automatically classifies the cleaned data according to a pre-set consumer complaint classification system, and manually labels key data, specifically including: A consumer complaint classification system with pre-classification algorithms based on common consumer complaint issues and areas; Unify the format of all types of collected data and remove noise information in the data; During the data cleaning process, duplicate data records are identified and removed; Extract features from the cleaned data, where the features include keywords, phrases, sentence structures, and sentiment tendencies in the text; The extracted features are input into the classification algorithm, and the classification algorithm automatically classifies according to the preset classification system and features; Manually label key data.

4. A consumer public opinion monitoring system according to claim 3, characterized in that: The text analysis module uses the sentiment analysis algorithm in natural language processing technology to judge the sentiment tendency of the complaint text, and adopts the topic model algorithm to automatically mine the main complaint topics and related keywords from a large number of complaint texts, including: Using a trained sentiment analysis model, wherein the sentiment analysis model is trained based on consumer complaint text data with clear sentiment labels; The preprocessed complaint text is input into the sentiment analysis model, and the sentiment analysis model outputs the corresponding sentiment category and sentiment intensity score; The preprocessed complaint text is input into the topic model algorithm, which automatically mines the topic structure of the text based on the probability statistics principle of the co-occurrence relationship of words in the text; Output the keywords corresponding to each topic and the distribution probability of each topic in the text data.

5. A consumer public opinion monitoring system according to claim 4, characterized in that: The data mining module is based on the time series analysis method to collect and analyze complaint data in different time periods, predict the development trend of complaint hotspots, and analyze the differences in complaint characteristics of different regions and different consumer groups by combining geographic information system technology and consumer portrait data, including: Select the appropriate time series model based on the characteristics of complaint data in different time periods; Use time series models to fit and predict complaint data and predict the development trend of complaint hotspots within a preset time period; Collect geographic data related to complaint data; Associating complaint data with geographic data to form a complaint dataset containing geographic information; Use GIS technology to visualize complaint datasets; Collect consumer portrait data; The consumer portrait data is associated with the complaint data set to analyze the differences in complaint characteristics of different consumer groups in different regions.

6. A consumer public opinion monitoring system according to claim 5, characterized in that: The data reporting module provides a variety of report templates for enterprise and government users to choose from, and automatically generates a complete consumer public opinion monitoring report based on the selected template and visualized data, and sends it in a timely manner according to the push method set by the user, including: Use online public opinion monitoring tools to monitor public opinion information related to consumer sentiment in real time; Screen and filter the collected public opinion information, extract and organize the information related to the monitored object, and classify and label it; Conduct in-depth analysis of the filtered information, analyze the hot issues of public opinion, as well as the development trends and changing patterns of public opinion; Automatically generate consumer public opinion monitoring reports based on the selected template and set parameters; According to the push method and time set by the user, the public opinion monitoring report is automatically generated and sent to the receiving end specified by the user.

7. A consumer public opinion monitoring system according to claim 6, characterized in that: The system also includes a system management and maintenance module, which is used to set different permission levels for enterprise and government users, regularly and automatically update data collection sources, perform regular data backup, and monitor the system's operating status in real time.

8. A method for monitoring consumer public opinion, characterized in that: Specifically include: Use web crawler technology to regularly crawl consumption-related data from the Internet. At the same time, establish a secure connection with the database within the enterprise or platform to directly extract consumer complaint records; Clean the collected data, automatically classify the cleaned data according to the pre-set consumer complaint classification system, and manually mark the key data; Use sentiment analysis algorithms in natural language processing technology to judge the sentiment tendency of complaint texts and distinguish positive, negative or neutral comments. At the same time, use topic model algorithms to automatically mine the main complaint topics and related keywords from a large number of complaint texts; Based on the time series analysis method, the complaint data in different time periods are counted and analyzed to reveal the development trend and change law of complaint hotspots. Combined with geographic information system technology, the differences in complaint characteristics in different regions are analyzed. At the same time, combined with consumer portrait data, the differences in complaint characteristics of different consumer groups are analyzed. Based on the selected template and visualized data, a complete consumer opinion and public opinion monitoring report is automatically generated, and the consumer opinion and public opinion monitoring report is sent to relevant users in a timely manner according to the push method set by the user.

9. A computing device, characterized in that It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for monitoring consumer public opinion and sentiment as described in claim 8 are implemented.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for monitoring consumer public opinion and sentiment as described in claim 8.

Images