Biasness identification of talk show's host by using twitter data

Biasness is defined as the inclined feelings of a person about any event. It can be positive or negative. Biasness identification of TV talk show's host is an attractive topic to look into due to the influence of these media groups upon the opinion making. We are using twitter tweets upon different TV hosts's talk shows to find biasness. Twitter is a widely used microblog service and social network, where people tweet mostly about current events. Here, we use sentiments analysis to judge the biasness of talk show's host. The nature of the tweets can be judged by the sentiments analysis. We use Support Vector Machine and Naïve Bayes over user's tweets for biasness identification of the host. Furthermore, emoticons are widely used in sentiments over social media. Therefore, we study the frequency of emoticons from the recent Twitter data set. Then, we carry out analysis to examine the biasness through emoticons and sentiment polarity.