Intelligent Text Mining to Sentiment Analysis of Online Reviews

In the prevailing days social networking sites helps people to connect easily across the world and gain knowledge and also share their interests. But, unfortunately in some cases these sites became a platform for cyber bullying. Cyber bullying is the act which causes emotional and psychological distress leading to depression, anxiety, fear and low selfesteem to the victims. Cyberbullying can be elucidated as usage of digital communication typically by sending messages to threaten, defame, harass or intimidate someone. Common social media platforms like twitter, facebook, instagram are exposed to cyberbullying which has become very common now-a-days. This can be reduced to an extent if such intimidating messages or comments are segregated. The process of classifying a sentence whether it is positive, negative or neutral is known as sentiment analysis. It helps in determining emotional tone behind a sentence. To classify these intimidating messages this paper proposes a hybrid classifier approach which classifies reviews into positive or negative. Experimental results show that the accuracy of the classifier for considered dataset is 89.36%.