SOME NOVEL MEASUREMENT ON EXPLORING LARGE DATA SETS BASED ON MULTI-VARIABLES MUTUAL INFORMATION THEORY
Yushan Jiang, Qingling Zhang, Chao Liu · 2013
Applying for information theory, we present a measure of dependence for three-variable relationships: the three variables maximal information coefficient (3D-MIC). It is a kind of maximal information-based nonparametric exploration (MINE) statistics for identifying and classifying relationships in large data sets. 3D-MIC generalized the MIC measurement. At the same time some optimal single axis partition algorithm (OSPA) is built to ensure the feasibility of the MIC measurement.