An Ensemble of Shallow and Deep Learning Algorithms for Vietnamese Sentiment Analysis

Abstract

Sentiment analysis also known as opinion mining refers to the use of natural language processing to systematically identify and categorize opinions expressed in a piece of text. Recently, deep learning with ensemble techniques has achieved state of the art results in sentiment analysis. However, this approach has not been studied for Vietnamese corpus. In this paper, we propose an ensemble method by combining traditional (shallow) and deep learning algorithms. We tested our method on three Vietnamese sentiment datasets. The Experimental results showed that these approaches improve the accuracy of sentiment classification when compared to both individual deep and shallow algorithms.