HASLab

3/8/2017

By Prof. Paulo J. Azevedo, HASLab, INESC TEC & Minho University.

Abstract. In this talk I will briefly describe the data mining process and take association rules mining as an example of pattern mining. I will make use of association rules algorithms to tackle the subgroup mining problem. In subgroup mining one tries to identify interesting subpopulations according to a specific property or constraint. Typically, interesting means deviating from the expected. Several examples of techniques and patterns to describe interesting subgroups will be put forward.

Keywords. Machine Learning, Data Mining, Subgroup Mining, Association Rules, Pattern Mining.

About the speaker. Prof. Paulo J. Azevedo got his PhD from Imperial College at University of London (1995) and is now an Assistant Professor (with habilitation) at the Department of Informatics University of Minho. He is a member of the High-Assurance Software Lab (HASLab) – INESC TEC L.A. His research interests are Knowledge Discovery in Databases (data mining) and Machine Learning, in particular association rules, subgroup discovery, motif discovery in time series, graph mining, bioinformatics and recommendation systems. His most recent research interests include Distribution Learning and Social Networks. He has participated and led several research projects funded by the Portuguese research foundation. He was a reviewer for research projects submitted to the French ANR. He was a PC member of several data mining conferences as DS, PKDD, ECML and IJCAI. He was also a reviewer for several data mining journals as Intelligent Data Analysis: An International Journal, Information Science and Springer Data Mining and Knowledge Discovery.