Application of Uranium Mineral Band Feature Sub-set Selection Based on Genetic Algorithm

Abstract

Analyses show that the absorption band position determines the type of mineral radically. The paper proposes a method of applying GA (Genetic Algorithm) to the selection of the uranium mineral band feature sub-set. First, on the fundamental of the correlation between feature-based metrics: information entropy, information gain, symmetrical uncertainty and type space, the GA which is a random search algorithm uses the four standards as fitness functions to select the best feature points. Then set three different sub-intervals, extend the best feature points to the best feature sub-sets. Finally, the best feature sub-sets are used for classification. Experiments show that information gain and symmetrical uncertainty that based on genetic algorithm are better than based on CFS in classification.