Sentiment Analisis on Web-based Reviews using Data Mining and Support Vector Machine

摘要

This work aims to use sentiment analysis techniques, data mining, text mining and natural language processing to indicate the polarity of texts using support vector machine. Weka software and a movie review database from Internet Movie Database - IMDb - were used. This work uses preprocessing filters and WRAPPER techniques and Support Vector Machine (SVM) for classification. It presents better results when compared to other preprocessing techniques used in sentiment analysis.