Target Sentiment Analysis Ensemble for Product Review Classification

Rhoda Ogutu, Richard Rimiru, Calvins Otieno

OAI: oai:igi-global.com:299382 • DOI: 10.4018/JITR.299382

Abstract— Machine learning can be used to provide systems the ability to automatically learn and improve from experiences without being explicitly programmed. It is fundamentally a multidisciplinary field that draws on results from Artificial intelligence, probability and statistics, information theory and analysis, among other fields that impact the field of Machine Learning. Ensemble methods are techniques that can be used to improve the predictive ability of a Machine Learning model. An ensemble comprises of individually trained classifiers whose predictions are combined when classifying instances. Some of the currently popular ensemble methods include Boosting, Bagging and Stacking. In this paper, we review these methods and demonstrate why ensembles can often perform better than single models. Additionally, some new experiments are presented to demonstrate the computational ability of Stacking approach.