Leveraging Affective Hash tags for Ranking Music Recommendations
To implement a recommendation for music, based on the content of microblog, to extract affective contextual information from hashtags contained in these tweets.
Existing system uses manual intervention and usage based suggestion. The recommendation of any service is based on the number of users who already requested similar services on the same demographic. The results were not always accurate, as they are not considering the individual users need on requirement. This systems assumes that the popular music played by majority of people were good for recommendation.
We propose a system where the contextual data of the user is considered for the recommendation. When a user tweets about his interested music, as hashtag, he also writes about his view. We try to analyze the sentiment of the content pertaining to the user. The sentiment value and the song the user is interested is considered together. The music is ranked based on the users view and their sentiment.