Target Sentiment Analysis Model with Naïve Bayes and Support Vector Machine for Product Review Classification

Abstract

Sentiment analysis has demonstrated that the automation and computational recognition of sentiments is possible and evolving with time, due to factors such as; emergence of new technological trends and the continued dynamic state of the human language as a form of communication. Sentiment analysis is therefore an Information extraction task that aims at obtaining private sentiments that can either be classified as positive or negative, toward a specific object or subject. However, social media platforms are marred with informal texts that make extraction and parsing of relevant information a problem for most systems and models. This can pose as a challenge to business enterprises, individuals or organizations seeking to make specific strategic decisions based on the available data. To overcome such inefficiencies, this research first proposes implementation of two classifier models on the basis of feature selection and extraction; and performance evaluation on sentiment classification of product reviews. The research will explore the use of a detailed pre-processing technique with the implementation of Naïve Bayes and SVM classifiers. The effect in terms of performance measure of such computational models, evaluation of how the models can be implemented within Social Listening application fields and Machine Learning approaches to Sentiment analysis; has formed grounds for this research. This paper is however intended to further evaluate the performance of Naïve Bayes and Support Vector Machine (SVM) classifiers with an intension of integrating the two classifiers, and creating an ensemble model.