Speaker: Karell Bertet (Université de la Rochelle)
Time: 9h30, Thursday, January 16, 2020
Location: Room 612, Building A6, Institute of Mathematics.
Abstract: With the explosion of cdata, both in size and type, the analysis of complex and heterogeneous data is a current challenge, and there are few tools to understand this data in all its diversity.
Galactic is a new pattern mining tools we have developed that constructs a concept lattice for complex and heterogeneous mining using a pattern discovery approach.
This tool implements our new Next Priority Concept algorithm that extends Formal Concept Analysis to non binary data. It is inspired from pattern structure theory, where data are locally described by predicates independent of their types, allowing the management of heterogeneous data.
In this talk, I will first present the basics of Formal Concept Analysis for binary data. Then, I will describe our Next Priority Concept algorithm and its implementation named Galactic, with a system of plugins for an easy integration of new characteristics and new description. |
