Date(s) - 25/02/2019
11 h 00 min - 12 h 00 min
Catégories Pas de Catégories
In many areas, huge amounts of data have been generated in recent decades that are no longer possible to analyze manually. Data mining aims to provide solutions to this problem by proposing methods for finding interesting structures or patterns within the data from which knowledge can be deduced.
In this presentation, I will discuss the problems of searching large amounts of data from a variety of sources. I will present new methods of data mining that jointly integrate spatial and temporal components.
In this work, the data is structured and takes the form of attributed trees or graphs. I will then present a method of integrating and mining heterogeneous data using distributional semantic concepts to predict the links between microRNAs and diseases. I will also discuss the optimization of this method using an evolutionary algorithm.