Low frequency keyword extraction with sentiment classification and cyberbully detection using fuzzy logic technique

Various kinds of audio and video data are generated everyday like chatting, blog posts and Twitter on wide range of products. Providing keywords for these audio files, thus allow the users to quickly obtain the gist of the lengthy recordings. Sentiment classification aims to detect information such as opinions, direct and indirect feelings expressed in text. In this proposed framework it will detect both Implicit and Explicit expressions and it will classify the Positive, Negative and Neutral words and also to identify the topic of the particular meeting transcripts. Cyberbullying is a social aggressive and it has powerful negative effects on individuals, specifically adolescents. In this proposed framework it additionally to detect the online social cruelty attack from meeting transcripts like Twitter, blog and Facebook also identify the low frequency keywords. Finally, the quality of the framework is going to improve using fuzzy logic technique.