An Empiric Study on Bangla Sentiment Analysis Using Hybrid Feature Extraction Techniques

In this modern technologically advanced world, Sentiment Analysis (SA) is a very important topic in every language due to its various trendy applications. But SA in Bangla language is still in a dearth level. This work focuses on examining different hybrid feature extraction techniques on Bangla SA using a new comprehensive dataset of 203,493 comments collected from various microblogging sites. In this study, we have implemented 21 different hybrid feature extraction methods including Bag of Words (BOW), N-gram, TF-IDF, TF-IDF-ICF, Word2Vec, FastText, GloVe, Bangla-BERT etc with CBOW and Skipgram mechanisms. The proposed novel method (Bangla-BERT+Skipgram) outperforms all other feature extraction techniques in machine leaning (ML), ensemble learning (EL) and deep learning (DL) approaches. The (Bangla-BERT+Skipgram) method achieved highest 92.37%, 92.55% and 95.71% accuracy in ML, EL and DL algorithms.