Evolutionary models in the Big Data context

by Isaac Triguero (University of Nottingham)

16:00 (60 min) in Daysh G.07

In the era of big data, the leverage of recent advances achieved in distributed technologies enables data mining techniques to discover unknown patterns or hidden relations from voluminous data in a faster way. However, the issues posed by (real-world) complex data go beyond computational complexity, and big data mining techniques are confronted with multiple challenges w.r.t. scalability, dimensionality, class-imbalance, structured data types and lack of annotated samples.

In this talk, I will briefly introduce the big data learning problem and the technologies I am working with to manage such amount of data. Next, I will present some of my recent publications on this topic, focusing on evolutionary-based models for instance selection/generation, feature selection and class-imbalance datasets.