Constrained co-clustering with non-negative matrix factorisation. (30th August 2012)

• Record Type: Journal Article
• Authors: Salunke, Amit; Liu, Xumin; Rege, Manjeet
• Is Part Of: International journal of business intelligence and data mining. Volume 7:Number 1/2(2012)
• Page Start: 60
• Page End: 79
• Publication Date: 2012-08-30
• Subjects: semi-supervised -- non-negative -- matrix -- factorisation -- clustering -- co-clustering -- constraint -- cannot-link -- must-link; 006.312
• Languages: English
• ISSNs: 1743-8187

Abstract

Co-clustering refers to the problem of deriving sub-matrices of the data matrix by simultaneously clustering the rows (data instances) and columns (features) of the matrix. While very effective in discovering useful knowledge, many of the co-clustering algorithms adopt a completely unsupervised approach. Integration of domain knowledge can guide the co-clustering process and greatly enhance the overall performance. We propose a semi-supervised Nonnegative Matrix-factorisation (SS-NMF) based framework to integrate domain knowledge in the form of must-link and cannot-link constraints. Specifically, we augment the data matrix by integrating the constraints using metric learning and then perform NMF to obtain co-clustering. Under the proposed framework, we present two approaches to integrate domain knowledge, viz. a distance metric learning approach and an information theoretic metric learning approach. Through experiments performed on real-world web service data and publicly available text datasets, we demonstrate the performance of the proposed SS-NMF based approach for data co-clustering.