Skip to main content
placeholder image

On the optimality of sequential forward feature selection using class separability measure

Conference Paper


Abstract


  • This paper studies sequential forward feature selection that uses the scatter-matrix-based class separability measure. We find that by adding a scale factor to each iteration of the conventional sequential selection, a sequential selection that guarantees the global optimum can be attained. We give a thorough theoretical proof of its optimality via a novel geometric interpretation, and this leads to a unified framework including the optimal sequential selection, the conventional sequential selection and the best-individual-N selection. In addition, we show that with our formulation, feature selection can be treated as a linear fractional maximization problem, and it can be efficiently solved by algorithms well developed in the literature. This gives a non-sequential globally optimal feature selection algorithm. Both theoretical and experimental study demonstrate their efficiency.

Authors


  •   Wang, Lei
  •   Shen, Chunhua (external author)
  •   Hartley, Richard (external author)

Publication Date


  • 2011

Citation


  • Wang, L., Shen, C. & Hartley, R. (2011). On the optimality of sequential forward feature selection using class separability measure. Proceedings of 2011 International Conference on Digital Image Computing: Techniques and Applications (DICTA) (pp. 203-208). USA: IEEE.

Scopus Eid


  • 2-s2.0-84863055744

Ro Metadata Url


  • http://ro.uow.edu.au/infopapers/3812

Has Global Citation Frequency


Start Page


  • 203

End Page


  • 208

Place Of Publication


  • USA

Abstract


  • This paper studies sequential forward feature selection that uses the scatter-matrix-based class separability measure. We find that by adding a scale factor to each iteration of the conventional sequential selection, a sequential selection that guarantees the global optimum can be attained. We give a thorough theoretical proof of its optimality via a novel geometric interpretation, and this leads to a unified framework including the optimal sequential selection, the conventional sequential selection and the best-individual-N selection. In addition, we show that with our formulation, feature selection can be treated as a linear fractional maximization problem, and it can be efficiently solved by algorithms well developed in the literature. This gives a non-sequential globally optimal feature selection algorithm. Both theoretical and experimental study demonstrate their efficiency.

Authors


  •   Wang, Lei
  •   Shen, Chunhua (external author)
  •   Hartley, Richard (external author)

Publication Date


  • 2011

Citation


  • Wang, L., Shen, C. & Hartley, R. (2011). On the optimality of sequential forward feature selection using class separability measure. Proceedings of 2011 International Conference on Digital Image Computing: Techniques and Applications (DICTA) (pp. 203-208). USA: IEEE.

Scopus Eid


  • 2-s2.0-84863055744

Ro Metadata Url


  • http://ro.uow.edu.au/infopapers/3812

Has Global Citation Frequency


Start Page


  • 203

End Page


  • 208

Place Of Publication


  • USA