Ensemble generation and feature selection for the identification of students with learning disabilities

摘要: In this paper, we have made an extensive study of artificial intelligence (AI) techniques like ensemble classifiers and feature selection for the identification students with learning disabilities. The experimental results show that our best method, which combines both selection, can correctly identify up to 50% disabilities (LD) 100% confidence. Also when predicting samples in ''junior high school'' using model built on ''elementary are used build predict dataset. particular, propose variants two recent Feature Transform-based methods (Rotation Forest Input Decimated Ensemble). Rotation Forest, set is randomly split into subsets Principal Component Analysis (PCA) transform features belong a subset. Ensemble first singles out given class i runs PCA data only. This transformation applied whole dataset classifier D"i trained these transformed patterns. choice limits size number classes. perform empirical comparison varying Transform method technique clustering overcome drawback Ensemble.