selfX: An open-source R package with autonomic abilities for data mining

Abstract

This article describes the key characteristics of the functionality implemented in the “selfX” R software package, developed under the European s-X-AIPI project as an open-source software and offered to the R community for use and enhancement. Specifically this package offers a range of more autonomic functions for both data exploration and model training activities, such as unsupervised and supervised feature selection, unsupervised anomaly detection and/or predictive algorithm selection. Moreover, it includes complementary functions for data cleaning, statistical tests and outlier and missing values treatment.