An Improved Lexicon Based Model for Efficient Sentiment Analysis on Movie Review Data

Every day a large set of data are collected for various purposes from different sources. Analyzing these large data sets is very essential as we do not need to use all the information depending on the application usage. Hence, mining this data set including text and sentiment is gradually becoming very important and useful for application purposes. Market analysis experts make plans for any production by taking into account the users’ feedback and buying habits. Using different sentiment analysis methods, these tasks can be accomplished successfully. In our research, we discuss here the existing lexicon analysis method and find out the limitations of its methodology, e.g., lower accuracy. Although some researchers have proposed and compared the accuracy of their models with the existing lexicon approaches, our proposed and developed customized model shows good accuracy for movie data reviews compared to the existing approaches.