Spring 2015 Seminar Series Dr. Alberto Cano University of Cordoba, Spain Title: Evolutionary Mining in Large Databases Abstract: The volume of data to analyze, stored in large databases is a challenging task for data mining algorithms, including those based on evolutionary computation. This presentation discusses and analyzes how data size and its characteristics affect the performance, accuracy and scalability of algorithms. Classifiers based on evolutionary computation are introduced to show the process of learning classification rules and their application to an educational data mining problem. Efficient and scalable solutions to classification problems based on massively parallel computing in graphic processing units (GPUs) are presented. Significant speedups are achieved when distributing the process into multiple GPU devices, which allows mining in very large databases within acceptable time. Biography : Alberto Cano received the Ph.D. degree in Computer Science from the University of Granada in 2014 and the MS degree in Soft Computing in 2011, from the same university. He also received the MS degree in Intelligent Systems from the University of Cordoba in 2013 and the BS degree in Computer Science in 2010. His research is currently funded by the Spanish Ministry of Education. He is member of the Knowledge Discovery and Intelligent Systems group and his research is focused on databases, soft computing, parallel computing and GPU computing. He has published 12 journal articles, 2 book chapters and 9 conference papers. In addition, he has published 15 papers on teaching practices and pedagogical innovation. When: 12:00pm-1:00pm, Monday, March 16, 2015 Where: Room E2221, School of Engineering-East Hall, Monroe Campus