Sentiment analysis for Turkish Twitter feeds

Sentiment analysis is one of the most useful tools in social media monitoring. Implementing sentiment analysis on data gained from social media (Blogs, Twitter, and Facebook) can increase the customer satisfaction and decrease the costs for a company. Also sentiment analysis can be used in various domains, such as economic, commercial and opinion mining for the users to get meaningful information. In this study, Turkish Twitter feeds collected from Twitter API have been analyzed in terms of the sentiment context whether positive or negative using document classification methods. Experimental results have been conducted on machine learning algorithms such as SVM, Naive Bayes, Multinomial Naive Bayes and KNN. The features represented by vector space are extracted from two different models which are Bag of Words and N-Gram. The experimental results have been investigated on the effect of classification methods.