Social network public opinion monitoring method and system based on artificial intelligence

Abstract

The invention discloses a social network public opinion monitoring method and system based on artificial intelligence. The method comprises the following steps of obtaining public opinion data from a social network platform and performing data preprocessing; hot events are automatically extracted from the public opinion data, and a density peak clustering algorithm is utilized to identify a high-density area and determine the popularity of the hot events. A BERT model is adopted to carry out sentiment classification on a text, context features are extracted in combination with bidirectional LSTM, and sentiment tendency is output through a Softmax classifier. For a specific target event, a multi-head attention mechanism and a convolutional neural network are used for specific target emotion classification. And performing noise reduction and decomposition on the time sequence data, constructing a plurality of basic prediction models, generating a combined prediction model in combination with weighted average, and monitoring and predicting the public opinion development trend in real time. Accurate public opinion analysis and decision support are provided for governments and enterprises, public opinion risks are reduced, social stability is maintained, and social construction is promoted.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a social network public opinion monitoring method and system based on artificial intelligence. Background Art

[0002] With the rapid development of the Internet and social media, social networks have become an important channel for people to obtain news, express opinions and participate in social affairs. According to data from the China Internet Network Information Center (CNNIC), the proportion of social media users who obtain news information is as high as 90.7%. However, the openness and easy spread of social networks have led to the spread of false information becoming more and more common. Only 25.7% of users will verify the authenticity of the information before forwarding it, and 60.3% of netizens directly forward false information without verification. This phenomenon not only contributes to the spread of false news, but also affects the public opinion and social stability to a certain extent.

[0003] Social networks have become an important source of the generation and spread of hot events. With the formation of communication influence, social media not only guides the follow-up reports of news websites and traditional media, but also promotes a wider wave of public opinion. To meet this challenge, the government and enterprises urgently need to establish an effective public opinion monitoring and analysis mechanism to timely identify and guide public sentiment and maintain social stability.

[0004] In this context, it is particularly important to use emerging technologies such as artificial intelligence, machine learning and big data to conduct sentiment analysis and trend prediction of hot events on social networks. Through in-depth analysis of public opinion data, it is possible to identify hot events that the public is concerned about, judge their emotional tendencies, and analyze the development context and transmission path of social emergencies.

[0005] Existing public opinion monitoring methods mainly rely on machine learning and deep learning technologies. Although these methods have improved the accuracy and efficiency of public opinion analysis to a certain extent, they still have deficiencies in data preprocessing, topic model coverage, and automation. In addition, current sentiment analysis methods are often limited to text data and lack comprehensive analysis of multimodal data (such as pictures, videos, etc.). Summary of the invention

[0006] In view of the defects of the prior art, the present invention provides a social network public opinion monitoring method and system based on artificial intelligence, which realizes the automatic extraction, sentiment classification and trend prediction of social network hot events by combining advanced algorithms such as density peak clustering, BERT pre-training model, bidirectional LSTM, multi-attention CNN, etc.

[0007] In order to achieve the above invention object, the technical solution adopted by the present invention is as follows:

[0008] A social network public opinion monitoring method based on artificial intelligence includes the following steps:

[0009] Step 1: Data preprocessing: Obtain public opinion data from social networking platforms, including text, pictures, comments, and forwarded information; perform Chinese word segmentation on text data, remove stop words and invalid information; extract key information, including people, places, time, and events; and use the BERT model to encode text data into word vectors.

[0010] Step 2: Automatic extraction of hot events: cluster analysis is performed on public opinion data to identify cluster centers in high-density areas; hot events in social networks are determined by calculating the heat of each cluster; and hot events are ranked by importance based on heat indices.

[0011] Step 3, sentiment classification: Use the BERT model to encode the preprocessed text into word vectors, and combine the text topic features for feature fusion; use the bidirectional LSTM model to extract the contextual features of the text and capture the context information; use the Softmax classifier to perform sentiment classification on the features output by the LSTM and output positive, negative or neutral sentiment tendencies.

[0012] Step 4: Specific target sentiment classification: Obtain the word vector of the text based on the BERT model; Use the multi-head attention mechanism to capture multiple important sentiment features in the text; Input the attention mechanism results into the CNN model for local feature extraction; Classify the sentiment polarity of the specific target through the Softmax classifier.

[0013] Step 5: Prediction of hot event trends: reduce noise and decompose time series data; build multiple basic prediction models to predict the decomposed modes; combine the results of multiple basic prediction models by weighted average to form a combined prediction model;

[0014] Step 6: Design an online prediction module and combine it with a combined prediction model to monitor hot event data in real time and predict future trends.

[0015] Furthermore, in the data preprocessing step, named entity recognition (NER) technology is used to extract key entity information such as people, places, time and events from the text.

[0016] Furthermore, in the step of automatically extracting hot events, the public opinion data is clustered based on a density peak clustering algorithm.

[0017] Furthermore, in the step of automatically extracting hot events, hot events are determined by microblog heat indexes.

[0018] Furthermore, in the hot event trend prediction step, a variational mode decomposition algorithm is used to reduce noise and decompose the time series data, and multiple basic prediction models are constructed based on the decomposed modes.

[0019] Furthermore, in the hot event trend prediction step, the whale optimization algorithm based on dynamic reverse learning and golden sine is combined to optimize the parameters of the prediction model.

[0020] The present invention also discloses a social network public opinion monitoring system based on artificial intelligence, which can be used to implement the above-mentioned social network public opinion monitoring method based on artificial intelligence, specifically including:

[0021] Data preprocessing module: used to obtain public opinion data from social networking platforms, including text, pictures, comments, and forwarding information. Perform Chinese word segmentation on text data, remove stop words and invalid information, extract key information, and use the BERT model for word vector encoding.

[0022] Automatic extraction module for hot events: cluster analysis of public opinion data, identify cluster centers in high-density areas, calculate the heat of each cluster, determine hot events in social networks, and rank them by importance based on heat indicators.

[0023] Sentiment classification module: The BERT model is used to encode the word vector of the text, and the text theme features are combined for feature fusion. The bidirectional LSTM model is used to extract the contextual features of the text, and the Softmax classifier is used for sentiment classification, outputting positive, negative or neutral sentiment tendencies.

[0024] Specific target sentiment classification module: Based on the BERT model, the word vector of the text is obtained, and the multi-head attention mechanism is used to capture multiple important sentiment features in the text. The attention mechanism results are input into the CNN model for local feature extraction, and the sentiment polarity of the specific target is classified through the Softmax classifier.

[0025] Hot Event Trend Forecast Module: De-noise and decompose time series data, build multiple basic forecasting models to forecast the decomposed modes. Combine the results of multiple basic forecasting models by weighted average to form a combined forecasting model.

[0026] Online prediction module: Combined with the combined prediction model, it monitors hot event data in real time, predicts future trends based on the model, and provides real-time support for decision-making.

[0027] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the above-mentioned method for monitoring public opinion in a social network based on artificial intelligence is implemented.

[0028] The present invention also discloses a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned social network public opinion monitoring method based on artificial intelligence is implemented.

[0029] Compared with the prior art, the advantages of the present invention are:

[0030] 1. Using the density peak clustering method, the automatic extraction of hot events in social networks is realized, which significantly improves the efficiency and accuracy of public opinion information extraction.

[0031] 2. A sentiment classification model based on BERT and bidirectional LSTM was built, which can accurately identify user emotional tendencies and improve the accuracy of sentiment classification.

[0032] 3. A specific target sentiment classification model based on BERT and multi-attention CNN was established, which improved the training efficiency and classification effect of the model and enhanced the ability to understand complex emotions.

[0033] 4. An online combined optimization and prediction system for hot event trends was designed to achieve real-time monitoring and prediction of public opinion development trends, improving the accuracy and real-time performance of the prediction.

[0034] 5. It provides decision-making support for public opinion analysis and response for governments and enterprises, which helps reduce public opinion risks, maintain social stability, and promote harmonious social development. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flow chart of a social network public opinion monitoring method based on artificial intelligence according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and examples.

[0037] like Figure 1 As shown, the present invention provides a social network public opinion monitoring method based on artificial intelligence, comprising the following steps:

[0038] Step 1: Data Preprocessing

[0039] Data collection: Mainstream social networking platforms (such as Weibo, WeChat, forums, etc.) collect a large amount of public opinion data, including text, pictures, comments, forwarding and other information.

[0040] Chinese word segmentation: Perform Chinese word segmentation on the collected social network texts. Commonly used tools such as Jieba word segmentation or THULAC segment sentences and remove invalid information such as stop words and punctuation marks.

[0041] Named Entity Recognition (NER): Use NER technology to identify key entities in the text and extract key information such as people, places, time, and events to provide support for subsequent hot event analysis.

[0042] Time information extraction: Extract time information from text and associate it with events to ensure that hot events can be sorted and analyzed according to the timeline, laying the foundation for trend prediction.

[0043] Word vectorization: Use the BERT pre-trained model to encode the processed text data into word vectors. The BERT model captures the contextual information of the text through a bidirectional Transformer and generates high-quality word vector representations to facilitate feature learning in subsequent models.

[0044] Step 2: Automatically extract hot events

[0045] Density Peak Clustering (DPC): Use the density peak clustering algorithm to cluster text data in social networks. The algorithm calculates the density of each data point, identifies the density peak area, selects the high-density area as the clustering center, and forms different clusters, including:

[0046] For each data point, calculate the local density of the point. The local density can be determined by counting the number of points whose distance from the point is less than a certain cutoff distance, or by using a Gaussian kernel function to smooth the density estimate. The cutoff distance is an important parameter that affects the granularity of the clustering results.

[0047] For each data point, identify the point closest to it among the points with higher local density than it and calculate the distance. This distance is used to measure how well the point is separated from the higher density areas. For the points with the highest density, its distance is defined as the maximum distance from all other points to ensure that these points are identified as potential cluster centers.

[0048] According to the local density and distance, those points with high local density and large distance from other high-density points are selected as cluster centers. These points usually form obvious "peaks" in the two-dimensional space of density and distance, representing potential cluster centers.

[0049] Each data point is assigned to the cluster to which the cluster center closest to it belongs. In this way, multiple clusters are formed, each of which usually corresponds to a potential hot event or topic.

[0050] Determine hot events: Analyze each cluster and determine the hot events corresponding to each cluster based on the popularity of Weibo (such as the number of reposts, comments, and likes) and other indicators, including:

[0051] Extract multiple popularity indicators for each Weibo post, such as the number of reposts, the number of comments, and the number of likes. Each indicator represents the popularity dimension of the Weibo, and the weight can be set to comprehensively calculate the overall popularity of each Weibo.

[0052] The weighted summation method is used to multiply the value of each heat index by its corresponding weight, and then the results are added together to calculate the comprehensive heat value of each Weibo post.

[0053] Summarize the popularity of microblog posts in each cluster and calculate the overall popularity of the cluster. You can choose to calculate the average or sum of all microblog popularity in the cluster, depending on the analysis objectives and data characteristics. The mean is suitable for comparing the average popularity of different clusters, and the sum is suitable for comparing the total influence of each cluster.

[0054] Event importance ranking: Sort the identified hot events and determine the impact of the events based on indicators such as popularity and spread. Importance ranking can help discover and analyze major public opinion events more quickly.

[0055] Step 3: Sentiment Classification

[0056] BERT-based word vector generation: The preprocessed text data is encoded into word vectors through the BERT model to generate context-related high-dimensional vector representations.

[0057] Feature fusion: The word vectors generated by BERT are fused with the topic features extracted from the text (such as entities and keywords recognized by NER) to form a more informative feature representation, which facilitates the sentiment classification model to understand the text semantics.

[0058] Bidirectional LSTM model: The fused feature vector is input into the bidirectional LSTM (BiLSTM) network. The bidirectional LSTM enhances the ability to understand complex emotional expressions by capturing the contextual information of the text.

[0059] Sentiment classification: Use the Softmax classifier to classify the features output by the LSTM network, and finally obtain the sentiment of the text. The classification category can be positive, negative or neutral.

[0060] Step 4: Target-specific Sentiment Classification

[0061] BERT word vector acquisition: For sentiment classification of a specific target event, the BERT model is first used to encode the text and generate a word vector.

[0062] Multi-attention mechanism: Use a multi-head attention mechanism to capture multiple important sentiment features in the text. The multi-head attention mechanism allows the model to focus on multiple aspects of the text at the same time, thereby improving the sentiment classification accuracy of a specific target.

[0063] Multi-attention CNN model: The results of multiple attention mechanisms are input into the convolutional neural network (CNN). CNN captures the local sentiment features in the text through the local perception mechanism and performs classification.

[0064] Sentiment polarity output: The output of the CNN model is processed using a Softmax classifier to obtain the sentiment polarity (positive, neutral, or negative) of a specific target event.

[0065] Step 5: Prediction of hot event trends

[0066] Time series data acquisition: Collect time series data for each hot event, such as the event's propagation speed, popularity changes, forwarding volume, and other dynamic data.

[0067] Variational Mode Decomposition (VMD): Use the variational mode decomposition algorithm to reduce noise and decompose time series data, decompose the original time series into modes of different frequencies, and eliminate noise interference.

[0068] Construction of basic prediction models: Based on the decomposed modes, multiple basic prediction models are constructed. Commonly used models include support vector regression (SVR), long short-term memory network (LSTM), gated recurrent unit (GRU), etc.

[0069] Improved Whale Optimization Algorithm (WOA): The Whale Optimization Algorithm improved by dynamic reverse learning and golden sine is used to optimize the parameters of the basic prediction model, find the optimal parameter combination, and improve the accuracy of the prediction model.

[0070] Combined prediction model: The results of multiple basic prediction models are combined through weighted average and other methods to form a combined prediction model. The combined model can integrate the advantages of multiple models and improve the overall prediction accuracy.

[0071] Online prediction and monitoring: Build an online prediction module to monitor the changes in popularity and development trends of hot events in real time, combine the combined prediction model to predict future trends, and provide real-time public opinion response support for governments and enterprises.

[0072] The structure of the online prediction module is as follows:

[0073] Data Collection Layer

[0074] Real-time data collection: Use API interfaces, web crawlers, or stream data processing tools to continuously obtain time series data from social networks. These data include the propagation speed of events, changes in popularity, and forwarding volume.

[0075] Data storage and management: Store the collected data in an efficient database (such as the NoSQL database MongoDB or the time series database InfluxDB) to support fast query and processing.

[0076] Data preprocessing layer

[0077] Data cleaning: Remove missing values, outliers, and noise to ensure data quality and consistency.

[0078] Variational Mode Decomposition (VMD): VMD algorithm is used to reduce noise and decompose time series data, decomposing the original sequence into modes of different frequencies and eliminating noise interference.

[0079] Model building layer

[0080] Basic prediction model: For each decomposed mode, multiple basic prediction models are constructed, such as support vector regression (SVR), long short-term memory network (LSTM), gated recurrent unit (GRU), etc.

[0081] Model parameter optimization: The Whale Optimization Algorithm (WOA) improved by dynamic reverse learning and golden sine is used to optimize the parameters of the basic prediction model and find the optimal parameter combination to improve the prediction accuracy.

[0082] Model Integration Layer

[0083] Combined forecasting model: The forecast results of basic forecasting models are combined through weighted average or other integration methods to form a comprehensive forecasting model. The combined model can take advantage of the advantages of multiple models and improve the overall forecasting accuracy and robustness.

[0084] Prediction and Monitoring Layer

[0085] Real-time prediction: Use the combined prediction model to make real-time predictions on future trends of hot events. The prediction results can include short-term and long-term predictions to meet different application needs.

[0086] Monitoring and Alerts: Set monitoring indicators and alert thresholds. When the heat change of a hot event exceeds the set range, an alarm is triggered to notify decision makers or relevant personnel.

[0087] Visualization and user interaction layer

[0088] Data visualization: Dynamically display the time series data, forecast results and change trends of hot events through charts, dashboards, etc., to facilitate user understanding and analysis.

[0089] User interaction: Provides a user interface that allows users to customize monitoring indicators, prediction cycles, and alarm thresholds to meet the needs of different users.

[0090] Construction ideas

[0091] Modular design: The online prediction module is divided into multiple independent functional layers, and data transmission and interaction are carried out between each layer through interfaces. This modular design facilitates the expansion and maintenance of the system.

[0092] Real-time and high efficiency: Use stream data processing technology and efficient algorithms to achieve real-time data collection and processing, ensuring that the prediction module can quickly respond to data changes and provide timely prediction results.

[0093] Algorithm integration and optimization: Combine multiple prediction algorithms to improve prediction accuracy through model integration. At the same time, use improved optimization algorithms to adjust model parameters to obtain the best prediction performance.

[0094] Through the above ideas and designs, the online prediction module can realize real-time monitoring and trend prediction of hot events in social networks. This can not only help decision makers understand the development of events in a timely manner, but also provide strong support for public opinion management and crisis response. The design of the module should fully consider the real-time, flexibility and security of the system to ensure efficient and reliable performance in practical applications.

[0095] In another embodiment of the present invention, a social network public opinion monitoring system based on artificial intelligence is provided. The system can be used to implement the above-mentioned social network public opinion monitoring method based on artificial intelligence, specifically including:

[0096] Data preprocessing module: used to obtain public opinion data from social networking platforms, including text, pictures, comments, and forwarding information. Perform Chinese word segmentation on text data, remove stop words and invalid information, extract key information, and use the BERT model for word vector encoding.

[0097] Automatic extraction module for hot events: cluster analysis of public opinion data, identify cluster centers in high-density areas, calculate the heat of each cluster, determine hot events in social networks, and rank them by importance based on heat indicators.

[0098] Sentiment classification module: The BERT model is used to encode the word vector of the text, and the text theme features are combined for feature fusion. The bidirectional LSTM model is used to extract the contextual features of the text, and the Softmax classifier is used for sentiment classification, outputting positive, negative or neutral sentiment tendencies.

[0099] Specific target sentiment classification module: Based on the BERT model, the word vector of the text is obtained, and the multi-head attention mechanism is used to capture multiple important sentiment features in the text. The attention mechanism results are input into the CNN model for local feature extraction, and the sentiment polarity of the specific target is classified through the Softmax classifier.

[0100] Hot Event Trend Forecast Module: De-noise and decompose time series data, build multiple basic forecasting models to forecast the decomposed modes. Combine the results of multiple basic forecasting models by weighted average to form a combined forecasting model.

[0101] Online prediction module: Combined with the combined prediction model, it monitors hot event data in real time, predicts future trends based on the model, and provides real-time support for decision-making.

[0102] In another embodiment of the present invention, a terminal device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding functions; the processor described in the embodiment of the present invention can be used for the operation of the social network public opinion monitoring method based on artificial intelligence.

[0103] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one disk storage.

[0104] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the social network public opinion monitoring method based on artificial intelligence in the above embodiment; one or more instructions in the computer-readable storage medium are loaded and executed by the processor.

[0105] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may 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.

[0106] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0107] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0108] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0109] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the implementation methods of the present invention, and should be understood that the protection scope of the present invention is not limited to such special statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A social network public opinion monitoring method based on artificial intelligence, characterized in that: The following steps are involved: Step 1: Data preprocessing: Obtain public opinion data from social networking platforms, including text, pictures, comments, and forwarding information; perform Chinese word segmentation on text data, remove stop words and invalid information; extract key information, including people, places, time, and events; use the BERT model to encode word vectors for text data; Step 2: Automatic extraction of hot events: cluster analysis is performed on public opinion data to identify cluster centers in high-density areas; hot events in social networks are determined by calculating the heat of each cluster; and hot events are ranked in importance according to heat indices; Step 3: Sentiment classification: Use the BERT model to encode the preprocessed text into word vectors, and perform feature fusion based on the text topic features; Use the bidirectional LSTM model to extract the contextual features of the text and capture the contextual information. Use the Softmax classifier to perform sentiment classification on the features output by the LSTM and output positive, negative or neutral sentiment tendencies. Step 4: Specific target sentiment classification: Obtain the word vector of the text based on the BERT model; Use the multi-head attention mechanism to capture multiple important sentiment features in the text; The attention mechanism results are input into the CNN model for local feature extraction; the sentiment polarity of specific targets is classified through the Softmax classifier; Step 5: Prediction of hot event trends: reduce noise and decompose time series data; build multiple basic prediction models to predict the decomposed modes; combine the results of multiple basic prediction models by weighted average to form a combined prediction model; Step 6: Design an online prediction module and combine it with a combined prediction model to monitor hot event data in real time and predict future trends.

2. The method for monitoring public opinion on social networks according to claim 1, characterized in that: In the data preprocessing step, named entity recognition technology is used to extract key entity information such as people, places, time and events from the text.

3. The method for monitoring public opinion on social networks according to claim 1, characterized in that: In the step of automatically extracting hot events, the public opinion data is clustered based on a density peak clustering algorithm.

4. The method for monitoring public opinion on a social network according to claim 1, characterized in that: In the step of automatically extracting hot events, hot events are determined by microblog heat indexes.

5. The method for monitoring public opinion on social networks according to claim 1, characterized in that: In the hot event trend prediction step, a variational mode decomposition algorithm is used to reduce noise and decompose time series data, and multiple basic prediction models are constructed based on the decomposed modes.

6. The method for monitoring public opinion on a social network according to claim 1, characterized in that: In the hot event trend prediction step, the parameters of the prediction model are optimized by combining the whale optimization algorithm based on dynamic reverse learning and golden sine.

7. A social network public opinion monitoring system based on artificial intelligence, characterized by: The system can be used to implement the social network public opinion monitoring method according to any one of claims 1 to 6, specifically including: Data preprocessing module: used to obtain public opinion data from social networking platforms, including text, pictures, comments, and forwarding information; perform Chinese word segmentation on text data, remove stop words and invalid information, extract key information, and use the BERT model for word vector encoding; Hot Event Automatic Extraction Module: Perform cluster analysis on public opinion data, identify cluster centers in high-density areas, calculate the heat of each cluster, determine hot events in social networks, and rank them according to heat indicators; Sentiment classification module: The BERT model is used to encode the word vector of the text, and the text theme features are combined for feature fusion; the bidirectional LSTM model is used to extract the contextual features of the text, and the Softmax classifier is used for sentiment classification, and the positive, negative or neutral sentiment tendency is output; Specific target sentiment classification module: Based on the BERT model, the word vector of the text is obtained, and the multi-head attention mechanism is used to capture multiple important sentiment features in the text; the attention mechanism results are input into the CNN model for local feature extraction, and the sentiment polarity of the specific target is classified through the Softmax classifier; Hot Event Trend Forecasting Module: De-noise and decompose time series data, build multiple basic forecasting models to forecast the decomposed modes; combine the results of multiple basic forecasting models by weighted average to form a combined forecasting model; Online prediction module: Combined with the combined prediction model, it monitors hot event data in real time, predicts future trends based on the model, and provides real-time support for decision-making.

8. A computer device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for monitoring public opinion in a social network as claimed in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the program is executed by a processor, the social network public opinion monitoring method described in one of claims 1 to 6 is implemented.

Images