Knowledge Discovery in Scalable Real-time Data Mining Systems

Abstract

This paper looks at the current state-of-the-art scalable real-time data miningsystems, and explores possible improvements to the automated knowledge discov-ery process through potential improvements in feature selection, use of clusteringalgorithms, and the information evaluation process, while still maintaining highscalability and real-time performance. A framework is designed and built to testthe system on real-world data.