Sentiment Analysis Using Machine Learning Approach

Abstract

Customers feedback is a valuable asset for businesses, that can be used in order to improve their performance. One of the fastest spreading areas today in computer science - Sentiment Analysis, helps to extract precious information from textual data, in order to identify the feeling of a statement. This research aims to build a classifier to predict customers’ satisfaction, based on Amazon reviews dataset, for different brands of mobile phones. The paper proposes a comparison between four text classification algorithms - Naïve Bayes, Support Vector Machine, Decision Tree and Random Forest, using different feature extraction techniques, such as Bag of words and TF-IDF. In addition, the models are evaluated using accuracy, precision, recall and F-score metrics. Our experiments revealed that Support Vector Machine achieves the best results and is very suitable for classification of the sentiment on product reviews.

Keywords

• sentiment analysis
• customer reviews
• machine learning
• text classification