Parallel Architectures, Algorithms and Programming, International Symposium on (2011)
Dec. 9, 2011 to Dec. 11, 2011
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/PAAP.2011.61
Spectral clustering is widely used in these years. Recently, methods that connect spectral clustering and semi-supervised clustering become popular. These methods improve the result through using constraint information in spectral clustering. Generally, there are two ways to select constrained information, one is random selection method and the other is active learning method. Here we focus on active learning methods. In this paper, we propose an active learning process, which considers the local and global information of dataset, and decide which constraint to choose by studying the change of eigenvectors.
X. Liu, L. Zong, H. Lin and X. Zhang, "Active Semi-supervised Spectral Clustering," Parallel Architectures, Algorithms and Programming, International Symposium on(PAAP), Tianjin, China, 2011, pp. 95-99.