期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：2
DOI：10.15680/ijircce.2015.0302096
出版社：S&S Publications
摘要：Sentiment analysis, also called opinion mining, is the field of study that analyzes people’s opinions,sentiments, evaluations, appraisals, attitudes, and emotions. It plays major important role to analysis identify theemotions of the user. Earlier numbers of the works have been performed to analyze sentiments from speech, text anddocuments, from this the songs plays a most important to sentiment analysis since the songs and mood are mutuallydependent to each other. Based on the selected song it becomes easy to find the mood of the listener, in future it will beused for recommendation systems. Songs are considered as a text file to perform a song it becomes imperative to findthe hidden meaning of the song for mining the sentiment and classify them accordingly. Each song is a mixture ofmoods. In order to perform this process the input songs files are preprocessed using the semantic matching based on theWordNet Graph Representation and mining can be done by Hidden markov model (HMM ) which classifies the topicsinto either two (positive/negative) or multiple (happy/angry/sad/...) classes. Topics mined by HMM can representmoods. For validation, we have used a dataset that consists of the different moods annotated by users of a particularwebsite.
