Multiple Graph Label Propagation by Sparse Integration

• Published in: IEEE transaction on neural networks and learning systems Vol. 24; no. 12; pp. 1999 - 2012
• Main Authors: Karasuyama, Masayuki, Mamitsuka, Hiroshi
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.12.2013; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Applied sciences; Artificial intelligence; Classification; Combinatorics; Combinatorics. Ordered structures; Computational complexity; Computer science; control theory; systems; Data processing. List processing. Character string processing; Exact sciences and technology; Graph theory; Graph-based semisupervised learning; Graphs; Information retrieval. Graph; label propagation; Labels; Laplace equations; Learning; Learning and adaptive systems; Linear programming; Mathematics; Memory organisation. Data processing; multiple graph integration; Neural networks; Noise measurement; Optimization; Propagation; Representations; Sciences and techniques of general use; Semisupervised learning; Software; Sparsity; Symmetric matrices; Theoretical computing; Performance evaluation; Information integration; multiple graph integration; label propagation; Labelling; Graph theory; sparsity; Optimization; Graph-based semisupervised learning; Information source; Semi-supervised learning; Multiplicity; Classification

Graph-based approaches have been most successful in semisupervised learning. In this paper, we focus on label propagation in graph-based semisupervised learning. One essential point of label propagation is that the performance is heavily affected by incorporating underlying manifold of given data in...