Sentiment Analysis on Twitter by using Machine Learning Technique

Apply

Sentiment Analysis on Twitter by using Machine Learning Technique

Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2016, Vol 4, Issue 5

Abstract

Due to the vast opinion of rich web resources such as discussion forum, review sites , blogs and news corpora on the market in digital form, a lot of analysis is focusing on the area of sentiment analysis. People are trying to develop a system that can identify and classify opinion or sentiment as represented in an electronic text. A correct method to predict sentiments could modify us, to extract opinions from the internet and predict online customer’s preferences, might prove valuable for economic and market research. Till now, there are few different issues predominating in this analysis, i.e, sentiment classification, feature based classification and handling negations. In this thesis, we try to estimate the twitter posts about electronic products like mobiles, laptops etc.we collect the tweets of mobile from twitter and preprocess the tweets, in preprocessing remove irrelevent words such as name, symbols etc. after preprocessing we catogarized the tweets in two parts opinion or opinionless, we compare each tweets to database of positive, negative and average words those tweets carry positive, negative and average words, is in opinion cateogity other's discarted. now we check the polarity of opinion tweets, count the polarity of tweets and divide in positive, negative and average category. Estimating the probablity of positive tweets we use Naive bayes and S.V.M and maximum entropy and compare results for better probablity result.

Authors and Affiliations

Neha Upadhyay, Prof. Angad Singh

Keywords

Tweets Sentiment Analysis Machine Learning Techniques Naive Bayes SVM Maximum Entropy Ensemble Classifier

EP ID EP22160
DOI -
Views 216
Downloads 5