Issue No.05 - May (2014 vol.26)
Khalid Benabdeslem , LIRIS, Univ. of Lyon 1, Lyon, France
Mohammed Hindawi , LIRIS, INSA of Lyon, Lyon, France
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TKDE.2013.86
This paper describes a three-level framework for semi-supervised feature selection. Most feature selection methods mainly focus on finding relevant features for optimizing high-dimensional data. In this paper, we show that the relevance requires two important procedures to provide an efficient feature selection in the semi-supervised context. The first one concerns the selection of pairwise constraints that can be extracted from the labeled part of data. The second procedure aims to reduce the redundancy that could be detected in the selected relevant features. For the relevance, we develop a filter approach based on a constrained Laplacian score. Finally, experimental results are provided to show the efficiency of our proposal in comparison with several representative methods.
learning (artificial intelligence), data reduction, feature selection,dimensionality reduction, efficient semi supervised feature selection, constraint, relevance, redundancy, three-level framework, high-dimensional data optimization, pairwise constraint selection, filter approach, constrained Laplacian score,Laplace equations, Feature extraction, Coherence, Redundancy, Context, Data mining, Vectors,Redundancy, Feature selection, semi-supervised learning, Constraints,redundancy, Semi-supervised feature selection, constraints, relevance
Khalid Benabdeslem, Mohammed Hindawi, "Efficient Semi-Supervised Feature Selection: Constraint, Relevance, and Redundancy", IEEE Transactions on Knowledge & Data Engineering, vol.26, no. 5, pp. 1131-1143, May 2014, doi:10.1109/TKDE.2013.86