An Enhanced Hybrid Feature Selection Technique Using Term Frequency-Inverse Document Frequency and Support Vector Machine-Recursive Feature Elimination for Sentiment Classification

Sentiment classification is increasingly used to automatically identify a positive or negative sentiment in text review. In classification, feature selection had always been critical and challenging problem. Most of the related for techniques unable overcome problems evaluating significant features that will reduce performance. This paper proposes an enhanced hybrid technique improve based on machine learning approaches. First, two customer review datasets namely Labelled large IMDB are retrieved pre-processed. Next, proposed which hybridization Term Frequency-Inverse Document Frequency (TF-IDF) Supports Vector Machine (SVM-RFE) developed tested these datasets. TF-IDF aims measure importance. The SVM-RFE iteratively evaluates ranks features. For only ktop from ranked be used. Finally, Support (SVM) classifier deployed observe performance technique. measured using accuracy, precision, recall, F-measure. experimental results show promising performances with 84.54% 89.56% measurements especially dataset. also outperformed other certain Consequently, able 19.25% 70.5% classified. reduction rate optimally utilizing computational resources while maintaining efficiency