Domain-Specific features clustering aims to cluster the features from related domains into K clusters. Although traditional clustering algorithms can be used to domain-specific features clustering, the performance may not good as the features have little inter-connection in related domains. In this paper, we develop a solution that uses the domain-independent feature as a bridge to connect the domain-specific features. And we use spectral clustering to cluster the domain-specific features into K clusters. We present theoretical analysis to show that our algorithm is able to produce high quality clusters. The experimental results show that our algorithm improves the clustering performance over the traditional algorithms. © 2012 Springer-Verlag.
CITATION STYLE
Yang, X., Wang, M., Fang, L., Yue, L., & Lv, Y. (2012). Research on domain-specific features clustering based spectral clustering. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7332 LNCS, pp. 84–92). https://doi.org/10.1007/978-3-642-31020-1_11
Mendeley helps you to discover research relevant for your work.