Mining sentiments from songs using latent dirichlet allocation

Sharma, Govind and Murty, Narasimha M (2011) Mining sentiments from songs using latent dirichlet allocation. In: IDA'11 Proceedings of the 10th international conference on Advances in intelligent data analysis X, 2011, Berlin, Heidelberg.

PDF ida2011.pdf - Published Version Restricted to Registered users only Download (207kB) | Request a copy

Abstract

Song-selection and mood are interdependent. If we capture a song’s sentiment, we can determine the mood of the listener, which can serve as a basis for recommendation systems. Songs are generally classified according to genres, which don’t entirely reflect sentiments. Thus, we require an unsupervised scheme to mine them. Sentiments are classified into either two (positive/negative) or multiple (happy/angry/sad/...) classes, depending on the application. We are interested in analyzing the feelings invoked by a song, involving multi-class sentiments. To mine the hidden sentimental structure behind a song, in terms of “topics”, we consider its lyrics and use Latent Dirichlet Allocation (LDA). Each song is a mixture of moods. Topics mined by LDA can represent moods. Thus we get a scheme of collecting similar-mood songs. For validation, we use a dataset of songs containing 6 moods annotated by users of a particular website.

Item Type:	Conference Proceedings
Series.:	LNCS
Publisher:	Springer-Verlag
Additional Information:	Copyright of this article belongs to Springer-Verlag.
Keywords:	Latent Dirichlet Allocation; Music Analysis; Sentiment Mining; Variational Inference
Department/Centre:	Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	15 Mar 2013 06:22
Last Modified:	15 Mar 2013 06:22
URI:	http://eprints.iisc.ac.in/id/eprint/45977

Actions (login required)

View Item