AITION: A scalable platform for interactive data mining

Abstract

AITION is a scalable, user-friendly, and interactive data mining (DM) platform, designed for analyzing large heterogeneous datasets. Implementing state-of-the-art machine learning algorithms, it successfully utilizes generative Probabilistic Graphical Models (PGMs) providing an integrated framework targeting feature selection, Knowledge Discovery (KD), and decision support. At the same time, it offers advanced capabilities for multi-scale data distribution representation, analysis & simulation, as well as, for identification and modelling of variable associations. AITION is built on top of Athena Distributed Processing (ADP) engine, a next generation data-flow language engine, capable of supporting large-scale KD on a variety of distributed platforms, such as, ad-hoc clusters, grids, or clouds. On the front end, it offers an interactive visual interface that allows users to explore the results of the KD process. The end result is that users not only understand the process that led to a statistical conclusion, but also the impact of that conclusion on their hypotheses. In the proposed demonstration, we will show AITION in action at various stages of the knowledge discovery process, showcasing its key features regarding interactivity and scalability against a variety of problems. © 2012 Springer-Verlag.