Speaker: Prof. Alberto Fernández Oliva (University of Havana, Cuba)
When: Thursday, November 24, 2016 — 3pm
Where: Sala seminari DISIM (Alan Turing Building)
Title: Rough Set applied to outliers detection
Abstract: Rough Set Theory plays an important role when reasoning is performed on imprecise data. This approach is based on the knowledge that an agent has about a certain reality and on their ability to discern phenomena, processes, objects, etc. In other words, it’s based on the capability of classifying data obtained from different sources. This theory has been applied effectively in many real life problems. If, on the one hand, the implementation of the Rough Sets Theory to the field of the KDD process has been occurring ever since its formulation by Z. Pawlak in the decade of 80’s, last century, in the past years the outliers detection has started to be seen as a KDD process with interests in itself. The combination of both approaches (Rough Sets as a foundation for the characterization and outliers detection), constitutes an absolutely new point of view, with a great potential for it’s the theoretical interest and its practical applicability.