Bridging Machine Learning Theory and Practice – What we have learned from participating ACM KDD CUP

Abstract

While it is possible to learn a variety of machine learning and data mining theories from lectures or books, applying them effectively and efficiently to the real-world data is a completely different story. Very often data miners have to suffer a painful process of trial and error in applying machine learning tools due to lack of experience. Dealing with the practical issues on data is rather an art than science. Nevertheless, in this talk I’ll share some of the experiences we have learned from participating ACM KDDCup for the past several years. Instead of theoretical machine learning techniques, this talk will be focused more on the practical issues and tricks that are important to train an effective and efficient classifier, based on several case studies about medical data mining, telcom user behavior mining, educational data mining, and music recommendation.