Key points are not available for this paper at this time.
The ever-growing World Wide Web (WWW) containing textual movie reviews from Facebook groups, personal blogs, newspapers, dedicated movie review sites has crucial impacts on intended users and review providers where users can find the desired movies in his preference by checking the reviews and review providers can build a better recommendation system that influences the marketing policy and revenues generated from advertising campaigns. The manual approach to analyze this textual reviews is complex and time-consuming and hence it requires specialized automated systems. Sentiment Analysis (SA) is such a tool which is the computational study of extracting opinions, emotions from textual data to build such automated systems and it has been implemented in many languages for movie review domain except Bangla; which is continuously increasing due to the emergence of Bangladeshi reviewers on WWW. Since none of the existing SA works addressed this domain, in this paper, we have developed a polarity detection system on textual movie reviews in Bangla by using two popular machine learning algorithms named Naive Bayes (NB) and Support Vector Machines (SVM) and provided a comparative results where SVM performed slightly better than NB by considering stemmed unigram as feature with an excellent precision of 0.86.
Banik et al. (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: