Analysis on Hinglish Opinion Using Multinomial Naive Bayes Algorithm

Abstract

Sentimental Analysis is the study that analyses people’s sentiments, appraisals, attitudes, and emotions towards entities such as products, services, organizations, individuals, issues, events, the topics, and their attributes. These days, the latest mobile devices and websites are interpreting mashup language based keyboards; this has enabled many users to express their opinions and views about products in ‘Hinglish’. This research is focused on conducting sentimental analysis of ‘Tweets’ written in Hinglish mashup language. The system built here is highly accurate as it is on Multinomial Naive Bayes algorithm that uses machine learning as basis to classify positive, negative or neutral opinion of products safety. The results obtained using that system has high true positive rate and low false rate as compared to previous Naive Bayes algorithm.