Feeling, emotions, views, and attitudes are all examples of sentiment. Because of the rapid
growth of the World Wide Web, People frequently express their feelings via social media,
blogs, ratings, and reviews on the internet. Due to the increase in textual data, it is
necessary to examine the concept of expressing sentiments and calculate insights for
business exploration. Sentiment analysis is frequently used by business owners and
advertising agencies to develop new business strategies and advertising campaigns. This
paper we examine the problem of document classification by sentiment. i.e. classify a
document as negative document or as a positive document. We find out the machine
learning algorithms (Naïve Bayes, rule based JRip and J48 trees based) preform quite
efficiently on tackling this problem. We conclude by discussing more features that may
make those algorithms perform even better than the results we report.
أسامة الهادي علوش | Osamah Alhadi Alloush | 407