Sentiment Analysis of Bengali Texts on Online Restaurant Reviews Using Multinomial Naïve Bayes

Multi-class Textual Emotion Categorization using Ensemble of Convolutional and Recurrent Neural Network

“…However, very few researchers have investigated this issue in under-resourced languages like Bengali. With around 265 million native speakers, Bengali is the 7th most spoken language [25]. A massive number of people communicate through virtual platforms using territorial languages (i.e., Bengali).…”

Bangla Text Sentiment Analysis Using Supervised Machine Learning with Extended Lexicon Dictionary

“…The study determined sentiments through only assigning feature words to positive and negative tags without considering POS tagger. In a recent paper [31], 80.48% accuracy was attained during 6-fold cross validation approach in multinomial Naive Bayes classification. The authors used polarity from given dataset as a target output without generating any sentiment from texts.…”

Classification of Textual Sentiment Using Ensemble Technique

“…Wahid et al [38] proposed a sentiment analysis model using LSTM to classify the Bengali text into positive, negative, and neutral classes with an accuracy of 95% over a dataset consists of 10,000 Facebook comments. Sharif et al [30] performed sentiment analysis on restaurant reviews which achieved an accuracy of 80.48% on 1000 reviews. Sarkar and Bhowmick [27] performed the sentiment analysis on the Bengali tweet dataset, where SVM and MNB classifiers were used for the classification.…”

“…A few studies attempted to analyze sentiment in Bengali using ML techniques (such as SVM, NB, DT, and RF). Most of these studies developed a classifier model on a small dataset and classification task performed on a specific domain such as book reviews [13], restaurant reviews [15,30], and social media status [16]. Thus, the previous systems suffer from lower accuracy and generability.…”