TY - GEN
T1 - Locality-preserving L1-graph and its application in clustering
AU - Han, Shuchu
AU - Huang, Hao
AU - Qin, Hong
AU - Yu, Dantong
N1 - Publisher Copyright:
Copyright 2015 ACM.
PY - 2015/4/13
Y1 - 2015/4/13
N2 - Constructing a good graph to represent data structures is critical for many important machine learning tasks such as clustering and classification. Recently, a nonparameteric graph construction method called L1-graph is proposed with claimed advantages on sparsity, robustness to data noise and datum-adaptive neighborhood. However, it suffers a lot from the loss of locality and the instability of perfor- mance. In this paper, we propose a Locality-Preserving L1-graph (LOP-L1), which preserves higher local-connections and at the same time maintains sparsity. Besides, compared with L1-graph and the succeeding regularization-based tech- niques, our LOP-L1 requires less amount of running time in the scalability test. We evaluate the effectiveness of LOP-L1 by applying it to clustering application, which confirms that the proposed algorithm outperforms related methods.
AB - Constructing a good graph to represent data structures is critical for many important machine learning tasks such as clustering and classification. Recently, a nonparameteric graph construction method called L1-graph is proposed with claimed advantages on sparsity, robustness to data noise and datum-adaptive neighborhood. However, it suffers a lot from the loss of locality and the instability of perfor- mance. In this paper, we propose a Locality-Preserving L1-graph (LOP-L1), which preserves higher local-connections and at the same time maintains sparsity. Besides, compared with L1-graph and the succeeding regularization-based tech- niques, our LOP-L1 requires less amount of running time in the scalability test. We evaluate the effectiveness of LOP-L1 by applying it to clustering application, which confirms that the proposed algorithm outperforms related methods.
KW - L-graph
KW - Locality
KW - Sparsity
UR - http://www.scopus.com/inward/record.url?scp=84955457768&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84955457768&partnerID=8YFLogxK
U2 - 10.1145/2695664.2695710
DO - 10.1145/2695664.2695710
M3 - Conference contribution
AN - SCOPUS:84955457768
T3 - Proceedings of the ACM Symposium on Applied Computing
SP - 813
EP - 818
BT - 2015 Symposium on Applied Computing, SAC 2015
A2 - Shin, Dongwan
PB - Association for Computing Machinery
T2 - 30th Annual ACM Symposium on Applied Computing, SAC 2015
Y2 - 13 April 2015 through 17 April 2015
ER -