Research on User Behaviour of Network Public Opinion Using Sentiment Analysis Algorithm

This research is aimed at analyzing the user behaviour of network public opinion using a novel sentiment analysis algorithm. Although social network platforms like Twitter have distinct features, including tweet size, misspellings, and unusual characters, sentiment evaluation on these platforms is essential, yet existing categorization techniques mostly target textual content. So, in this research, an artificial algae-optimized adaptable support vector machine (AAO-ASVM) approach is proposed. The AAO method is applied to enhance the performance of the ASVM regarding effective sentiment analysis. Initially, we gather social network data samples, like Twitter, from a public source to train the proposed method. The gathered data samples are pre-processed for preparing and filtering textual data. This is followed by the presentation of the feature extraction technique known as term frequency-inverse document frequency (TF-IDF). From the extracted features, the proposed method is applied in sentiment analysis to analyze user behaviour in network public opinion. This research is developed on the Python platform to analyze the proposed method's performance regarding sentiment analysis. From the experimented outcomes, it can be concluded that the AAO-ASVM approach achieves the maximum performance in sentiment analysis compared to other existing studies.