Sentiment Analysis of Malayalam film review using machine learning techniques

• Published in: 2015 International Conference on Advances in Computing, Communications and Informatics (ICACCI) pp. 2381 - 2384
• Main Authors: Nair, Deepu S., Jayan, Jisha P., Rajeev, R. R., Sherly, Elizabeth
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2015
• Subjects: Accuracy; Conditional Random Field; Context; Motion pictures; Semantics; Sentiment analysis; Support Vector Machine; Support vector machines; Training
• ISBN: 9781479987900; 1479987905
• DOI: 10.1109/ICACCI.2015.7275974

The unprecedented growth of data in web, social media and the attempt to make the cognitive process using computers make Sentiment Analysis a challenging and interesting research problem. Sentiment Analysis mainly deals with the process of analyzing the sentiments or feelings from someone's expression or piece of information, and also in discovering the cognitive behavior of humans. The usage of computers to get feedback, opinion or remarks about a product, entertainment or political view of the public is very common. This paper demonstrates how Sentiment Analysis can be used in reviewing Malayalam films by using machine learning techniques. It is a hybrid approach comprising of machine learning techniques and rule based approach. This work would help the users to analyze the film criticism and also to assign the rank and popularity of new arrival films. In this work, the system checks the polarity at the sentence level, resulted in an accuracy of 91%.