Maximal information coefficient for feature selection for clinical document classification
Published in ICML workshop on machine learning for clinical data. Edingburgh, UK, 2012
Recommended citation: Chen Lin, Timothy Miller, Dmitriy Dligach, R Plenge, E Karlson, and G Savova. 2012. Maximal information coefficient for feature selection for clinical document classification. In ICML workshop on machine learning for clinical data. Edingburgh, UK. https://people.cs.pitt.edu/~milos/icml_clinicaldata_2012/Papers/Poster_Chen_Guergana_ICML_Clinical_2012.pdf
Abstract:
Abstractď€ Maximal Information Coefficient (MIC) is a novel correlation statistic that measures the association strength of linear and non-linear relationships between paired variables. We describe our first attempt in applying MIC in the clinical domain for a textual feature evaluation. The effect of MIC is compared to the Pearson Correlation-based feature selection method.