A Comparative Study of Ensemble Learning Methodologies for Multilingual Sentiment Analysis
Abstract
The widespread use of multiple social media platforms has amplified the expression of public opinions over the Internet in languages such as English, Spanish and many more. With the aid of technological advancements in machine learning, we can analyze reviews posted on the Internet and gauge public sentiments. There are organizations and businesses that are interested in the evaluation of these sentiments as this data can be generally used to obtain an opinion about a product, a restaurant, a candidate, etc. In this study, we perform a comparative analysis of three popular ensemble learning methodologies (Bagging, Boosting and Stacking) based on multiple base learner models like Support Vector Machines, Convolutional Neural Networks, Recurrent Neural Networks, Gated Recurrent Units and Bidirectional Long Short Term Memory for sentiment analysis. Based on the comparative study, the results show that the stacking and bagging ensembles performed comparatively better than the boosting ensembles to identify the correct sentiment based on the multilingual data.
