Using covariates for improving the minimum Redundancy Maximum Relevance feature selection method


Kursun O., Sakar C. O., Favorov O., AYDIN N., Gurgen F.

TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, cilt.18, sa.6, ss.975-987, 2010 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 18 Sayı: 6
  • Basım Tarihi: 2010
  • Doi Numarası: 10.3906/elk-0906-75
  • Dergi Adı: TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.975-987
  • Anahtar Kelimeler: Mutual information, mRMR, unsupervised learning, support vector machines, SINBAD covariates, MUTUAL INFORMATION, NEURAL-NETWORKS, PREDICTION, VARIABLES
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

Maximizing the joint dependency with a minimum size of variables is generally the main task of feature selection. For obtaining a minimal subset, while trying to maximize the joint dependency with the target variable, the redundancy among selected variables must be reduced to a minimum. In this paper, we propose a method based on recently popular minimum Redundancy-Maximum Relevance (mRMR) criterion. The experimental results show that instead of feeding the features themselves into mRMR, feeding the covariates improves the feature selection capability and provides more expressive variable subsets.