Paper Info
Title
Class-Dependent Locality Preserving Projections for Multimodal Scenarios
Abstract
This paper proposes a method for linear feature extraction called Class-dependent Locality Preserving Projections. It is a supervised extension of the Locality Preserving Projection algorithm and it aims to work in scenarios with within-class multimodality, which are those scenarios where the scattering of the patterns follows more than one modal distribution. Differently from the classical feature extraction techniques that build their solutions based on the whole dataset, the Class-dependent Locality Preserving Projections looks at each class separately, building a specific projection for each class. The proposed technique analyses a query pattern based on the output of each class and chooses the class that better fit the pattern. The experimental study shows that the Class-dependent Locality Preserving Projections is a feature extraction technique for general purposes, however, it is particularly well succeed when applied to within-class multimodal scenarios.
Year
DOI
Venue
2012
10.1109/ICTAI.2012.139
ICTAI
Keywords
Field
DocType
feature extraction technique,class-dependent locality preserving projections,proposed technique,within-class multimodality,experimental study,within-class multimodal scenario,multimodal scenarios,query pattern,linear feature extraction,classical feature extraction technique,locality preserving projection algorithm,principal component analysis,feature extraction,data analysis
Data mining,Multimodality,Locality,Dimensionality reduction,Dykstra's projection algorithm,Pattern recognition,Computer science,Feature extraction,Artificial intelligence,Modal,Machine learning,Principal component analysis
Conference
Volume
ISSN
ISBN
1
1082-3409
978-1-4799-0227-9
Citations 
PageRank 
References 
1
0.36
9
Authors
4
Name
Order
Citations
PageRank
Elias R. Silva Jr110.70
George D. C. Cavalcanti245152.60
Tsang Ing Ren320330.47
Silva, E.R.410.36