Abstract
Before clustering, we usually have some background knowledge about the data structure. Pairwise constraints are commonly used background knowledge. For graph partition problems, pairwise constraints can be naturally added to the graph edge. This paper integrates pairwise constraints into the objective function of graph cuts and derive the semi-supervised approximate spectral clustering based on Hidden Markov Random Fields (HMRF). This algorithm utilize the mathematical connection between HMRF semi-supervised clustering and approximate weighted kernel k-means. The approximate weighted kernel k-means is used to calculate the optimal clustering results of HMRF spectral clustering. The effectiveness of the proposed algorithm is verified on several benchmark data sets. Experiments show that adding more pairwise constraints will help improve the clustering performance. Our method has advantages for the challenging clustering tasks of large-scale nonlinear data because of the high efficiency and less memory consumption.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.