Term-weighting approaches in automatic text retrieval
Information Processing and Management: an International Journal
Introduction to Data Mining, (First Edition)
Introduction to Data Mining, (First Edition)
Integrating Features from Different Sources for Music Information Retrieval
ICDM '06 Proceedings of the Sixth International Conference on Data Mining
Multimodal Music Mood Classification Using Audio and Lyrics
ICMLA '08 Proceedings of the 2008 Seventh International Conference on Machine Learning and Applications
A document clustering algorithm for discovering and describing topics
Pattern Recognition Letters
Integration of text and audio features for genre classification in music information retrieval
ECIR'07 Proceedings of the 29th European conference on IR research
Improving mood classification in music digital libraries by combining lyrics and audio
Proceedings of the 10th annual joint conference on Digital libraries
Large Scale Semantic Access to Content (Text, Image, Video, and Sound)
Unsupervised music genre classification with a model-based approach
EPIA'11 Proceedings of the 15th Portugese conference on Progress in artificial intelligence
A Survey of Audio-Based Music Classification and Annotation
IEEE Transactions on Multimedia
| Hi-index | 0.00 |
In this paper an approach for music clustering, using only lyrics features, is developed for identifying groups with similar feelings, content or emotions in the songs. For this study, a collection of 30.000 Spanish lyrics has been used. The songs were represented in a vector space model (Bag Of Words (BOW)), and some techniques of Part Of Speech (POS) were used as part of preprocessing. Partitional and hierarchical methods were used to perform clustering estimating the appropriate number of clusters (k). For evaluating the clustering results, some internal measures were used such as Davies Bouldin Index (DBI), intra similarity and inter similarity measures. At last, the final clusters were tagged using top words and association rules. Experiments show that music could be organized in related groups and tagged using unsupervised techniques as clustering with only lyrics information.