Accés ràpid intranet

Més informació...

a a a



Finding frequent and rare itemsets using symbolic data mining methods


Laszlo Szathmary

Professor/a organitzador/a

Domnec Puig Valls


University of Debrecen, Hungary


06-07-2016 13:00


In data mining, frequent itemsets (FIs) and association rules play an important role. Due to the high number of patterns, various concise representations of FIs have been proposed, of which the most well-known representations are the frequent generators (FGs) and the frequent closed itemsets (FCIs). We present some vertical algorithms that are designed to extract FGs and FCIs from a dataset. Using the output of these algorithms, one can generate the set of minimal non-redundant association rules very easily. Frequent itemsets have been studied in detail in the literature. However, in some cases, rare itemsets and rare association rules can also contain important information just as frequent itemsets do. A particularly relevant field for rare itemsets is medical diagnosis for instance. At the end of the seminar we will show briefly our software toolkit called Coron, which incorporates a rich collection of data mining algorithms. Short bio: Dr. Laszlo Szathmary is an associate professor at the University of Debrecen, Hungary. He obtained his PhD in France, and then he was a postdoctoral research fellow in Montreal, Canada. His main research interests are data mining, formal concept analysis and its applications.


Laboratori 231