Abstract :
Sentiment analysis is basically analysing of the sentiments from the text. Sentiment analysis can be referred
as opinion mining. Sentiment analysis finds and justifies the sentiment of the person with respect to a given source
of content. The growth in micro-blogging activity on sites over the last few years has been increasing. Microblogging
sites like twitter contain a large amount of data, which helps the companies to know what public is thinking of them.
Sentiment analysis of this large data is very useful to express the opinion of the group of people. Twitter sentiment
analysis is tricky as compared to broad sentiment analysis because of the slang words and misspellings and repeated
characters. So it is very important to identify exact sentiment of each word. In our paper to obtain a highly accurate
model of sentiment analysis of tweets with respect to latest reviews of Games which include both mobile and PC.
With the help of feature selection and classifiers such as Support vector machine (SVM), Naïve Bayes and Maximum
Entropy. By adding feature selection to the classifier we can select the relevant features. We will classify these tweets
as positive, negative and neutral to give sentiment of each tweet and compare their results based on appropriate
models.
Keyword :
Sentiment analysis, opinion mining, micro-blogging, Support vector machine (SVM), Naive Bayes, Maximum Entropy.