TY - JOUR
T1 - Local adaptive projection framework for feature selection of labeled and unlabeled data
AU - Chen, Xiaojun
AU - Yuan, Guowen
AU - Wang, Wenting
AU - Nie, Feiping
AU - Chang, Xiaojun
AU - Huang, Joshua Zhexue
N1 - Publisher Copyright:
© 2012 IEEE.
PY - 2018/12
Y1 - 2018/12
N2 - Most feature selection methods first compute a similarity matrix by assigning a fixed value to pairs of objects in the whole data or to pairs of objects in a class or by computing the similarity between two objects from the original data. The similarity matrix is fixed as a constant in the subsequent feature selection process. However, the similarities computed from the original data may be unreliable, because they are affected by noise features. Moreover, the local structure within classes cannot be recovered if the similarities between the pairs of objects in a class are equal. In this paper, we propose a novel local adaptive projection (LAP) framework. Instead of computing fixed similarities before performing feature selection, LAP simultaneously learns an adaptive similarity matrix S and a projection matrix W with an iterative method. In each iteration, S is computed from the projected distance with the learned W and W is computed with the learned S. Therefore, LAP can learn better projection matrix W by weakening the effect of noise features with the adaptive similarity matrix. A supervised feature selection with LAP (SLAP) method and an unsupervised feature selection with LAP (ULAP) method are proposed. Experimental results on eight data sets show the superiority of SLAP compared with seven supervised feature selection methods and the superiority of ULAP compared with five unsupervised feature selection methods.
AB - Most feature selection methods first compute a similarity matrix by assigning a fixed value to pairs of objects in the whole data or to pairs of objects in a class or by computing the similarity between two objects from the original data. The similarity matrix is fixed as a constant in the subsequent feature selection process. However, the similarities computed from the original data may be unreliable, because they are affected by noise features. Moreover, the local structure within classes cannot be recovered if the similarities between the pairs of objects in a class are equal. In this paper, we propose a novel local adaptive projection (LAP) framework. Instead of computing fixed similarities before performing feature selection, LAP simultaneously learns an adaptive similarity matrix S and a projection matrix W with an iterative method. In each iteration, S is computed from the projected distance with the learned W and W is computed with the learned S. Therefore, LAP can learn better projection matrix W by weakening the effect of noise features with the adaptive similarity matrix. A supervised feature selection with LAP (SLAP) method and an unsupervised feature selection with LAP (ULAP) method are proposed. Experimental results on eight data sets show the superiority of SLAP compared with seven supervised feature selection methods and the superiority of ULAP compared with five unsupervised feature selection methods.
KW - Local structure learning
KW - sparse feature selection
KW - supervised feature selection
KW - unsupervised feature selection
UR - http://www.scopus.com/inward/record.url?scp=85047208343&partnerID=8YFLogxK
U2 - 10.1109/TNNLS.2018.2830186
DO - 10.1109/TNNLS.2018.2830186
M3 - 文章
C2 - 29994271
AN - SCOPUS:85047208343
SN - 2162-237X
VL - 29
SP - 6362
EP - 6373
JO - IEEE Transactions on Neural Networks and Learning Systems
JF - IEEE Transactions on Neural Networks and Learning Systems
IS - 12
M1 - 8361067
ER -