Filtering political sentiment in social media from textual information

Social media is now playing a vital role in influencing people's sentiment in favor or against a government or an organization. Therefore, to understand the sentiment of any posting in social media, an efficient method is an ultimate necessity. We have analyzed some facebook postings to understand political sentiments. In any politically motivated posting there are some dominant keywords. At first, we have prepared a dictionary consisting of unique words collected from political or nonpolitical posts or comments and then trained using Naïve Bayes algorithm based on probability theory. To identify the sentiment expressed in a new post or comment, we have extracted each word of the posting and then matched those with the dictionary words for classification. Finally, we have tested our algorithm using 200 postings from facebook and our result shows that the method can classify posts or comments with good accuracy.