Fuzzy Attention Based Deep Model for Sentiment Analysis

Monitoring user feedback and gratification is a critical concern for any business organization in the competitive world of internet businesses. Web forums and social media have developed into fantastic platforms in recent years for expressing or sharing thoughts about any good or interesting subject.The motive of sentiment analysis is to determine the emotions expressed in a text. A statement may be interpreted positively, neutrally, or negatively. But the truth is that each line does not necessarily express positivity, negativity, or neutrality in a straightforward manner. In order to create sentiment analysis findings that are more accurate, fuzzy sentiment analysis utilizing attention based CNN-RNN architecture is introduced. The model presented in the paper makes use of machine learning methods and a fuzzy approach for sentiment analysis and sentiment categorization on textual reviews and tweets. The suggested method was evaluated using real-world data, and the findings demonstrate that it improves tweet and review sentiment analysis effectiveness both in reference to accuracy and F1 score.