Revista UIS Ingenierías

Vol. 14 No. 1 (2015): Revista UIS Ingenierías
Articles

Sequential feature analysis in a floating search evaluation and extraction of weak metaclassifiers

PDF (Español (España)) HTML (Español (España))
  • Edwin Alberto Silva Cruz,
  • Carlos Humberto Esparza Franco
Edwin Alberto Silva Cruz
Universidad Industrial de Santander
Bio
Carlos Humberto Esparza Franco
Universidad Industrial de Santander
Bio

Published 2014-11-22

Keywords

  • Feature Search,
  • pattern recognition,
  • data mining,
  • dimensionality reduction,
  • classification system

How to Cite

Silva Cruz, E. A., & Esparza Franco, C. H. (2014). Sequential feature analysis in a floating search evaluation and extraction of weak metaclassifiers. Revista UIS Ingenierías, 14(1), 45–57. Retrieved from https://revistas.uis.edu.co/index.php/revistauisingenierias/article/view/45-57
Copyright Notice

Abstract

Feature extraction is one of the most challenging tasks in the design of a classification system. In this work we present a novel floating evaluation and search algorithm focused on weak features. In classification problem with a high number of weak features an exhaustive feature selection protocol is calculation cost prohibitive, so in our approach a floating method is proposed with restricted feature subset evaluation. Our proposal considerably decreases the calculation costs of feature search compared with conventional bottom-up, top-down and floating techniques, as well with other recent techniques, without reducing the classification performance. The proposed methodology was tested for 7-class facial expression recognition and the results show the viability of the approach for multiclass problems with weak features.

PDF (Español (España)) HTML (Español (España))

Downloads

Download data is not yet available.

References

  1. GUYON, I Feature extraction: foundations and applications.Springer, 2006.
  2. DEVIJVER, P; KITTLER, J. Pattern recognition:
  3. A statistical approach. Prentice/Hall International
  4. Englewood Cliffs, NJ, 1982.
  5. NAKARIYAKUL, S; CASASENT, D “An
  6. improvement on foating search algorithms for feature
  7. subset selection,” Pattern Recognit., vol. 42, no. 9, pp.
  8. –1940, 2009.
  9. PENG, Y; WU, Z; JIANG, J.“A novel feature selection
  10. approach for biomedical data classifcation,” J. Biomed.
  11. Inform., vol. 43, no. 1, pp. 15–23, 2010.
  12. GHEYAS, I; SMITH, L. “Feature subset selection in
  13. large dimensionality domains,” Pattern Recognit., vol.
  14. , no. 1, pp. 5–13, 2010.
  15. Z. Q; LU, J.“The elements of statistical learning: data
  16. mining, inference, and prediction,” J. R. Stat. Soc. Ser.
  17. A (Statistics Soc., vol. 173, no. 3, pp. 693–694, 2010.
  18. TRUNK, G. “A problem of dimensionality: A simple
  19. example,” Pattern Anal. Mach. Intell. IEEE Trans., no.
  20. , pp. 306–307, 1979.
  21. AHA, D; BANKERT, R. “A comparative evaluation of
  22. sequential feature selection algorithms,” in Learning
  23. from Data. Springer New York, 1996, pp. 199–206.
  24. WEDD, A; COPSEY, K. Statistical Pattern Recognition,
  25. third Edition, Third Edit. Wiley, 2011, p. 642.
  26. SUN, D; ZHANG, D.“Bagging constraint score for
  27. feature selection with pairwise constraints,” Pattern
  28. Recognit., vol. 43, no. 6, pp. 2106–2118, 2010.
  29. GU, Q. et al. “Generalized fsher score for feature
  30. selection,” arXiv Prepr. arXiv1202.3725, 2012.
  31. HANCAZAR, B. et al. “Small-sample precision of
  32. ROC-related estimates,” Bioinformatics, vol. 26, no. 6,
  33. pp. 822–830, 2010.
  34. PUDIL, P. et al. “Floating search methods in feature
  35. selection,” Pattern Recognit. Lett., vol. 15, no. 11, pp.
  36. –1125, 1994.
  37. SOMOL, P. et al.“Adaptive foating search methods in
  38. feature selection,” Pattern Recognigion Lett., vol. 20,
  39. no. 11, pp. 1157–1163, 1999.
  40. TIAN, Y. et al. “Recognizing Facial Actions by
  41. Combining Geometric Features and Regional
  42. Appearance Patterns,” Robot. Institute, Carnegie
  43. Mellon Univ. Pittsburgh, PA 15213, p. 31, 2001.
  44. LUCEY, P. et al. “The Extended Cohn-Kanade
  45. Dataset (CK+): A complete dataset for action unit and
  46. emotion-specifed expression,” in Computer Vision and
  47. Pattern Recognition Workshops (CVPRW), 2010 IEEE
  48. Computer Society Conference on, 2010, no. July, pp.
  49. –101.
  50. VU, N-S; CAPLIER, A.“Face recognition with patterns
  51. of oriented edge magnitudes,” in Computer Vision--
  52. ECCV 2010, Springer, 2010, pp. 313–326.
  53. SILVA, E. et al. “POEM-based facial expression
  54. recognition, a new approach,” in Image, Signal Processing, and Artifcial Vision (STSIVA), 2012 XVII
  55. Symposium of, 2012, pp. 162–167.
  56. JACCARD, P. Coeffcient: Jaccard, Étude comparative
  57. de la distribution forale dans une portion des Alpes et
  58. des Jura, vol. 37. Impr. Corbaz, 1901.
  59. CHA, S-H. “Comprehensive survey on distance/
  60. similarity measures between probability density
  61. functions,” City, vol. 1, no. 2, 2007.
  62. YU, L; LIU, H.“Feature selection for high-dimensional
  63. data: A fast correlation-based flter solution,” in ICML,
  64. , vol. 3, pp. 856–863.
  65. PRESS, W. et al. Flannery, Numerical recipes in C, vol.
  66. Citeseer, 1996.
  67. DENG, S. et al. “A feature-selection algorithm based on
  68. Support Vector Machine-Multiclass for hyperspectral
  69. visible spectral analysis,” J. Food Eng., vol. 119, no. 1,
  70. pp. 159–166, 2013.

Most read articles by the same author(s)