Morisita-Based Feature Selection for Regression Problems

Data acquisition, storage and management have been im- proved, while the factors of many phenomena are not well known. Conse- quently, irrelevant and redundant features artificially increase the size of datasets, which complicates learning tasks, such as regression. To address this problem, feature selection methods have been proposed. This research introduces a new supervised filter based on the Morisita estimator of in- trinsic dimension. The algorithm is simple and does not rely on arbitrary parameters. It is applied to both synthetic and real data and a comparison with a wrapper based on extreme learning machine is conducted.