Trends and Challenges in the Industrial Applications of KDD

Abstract

As an applications driven field, Knowledge Discovery in Databases and Data Mining (KDD) techniques have made considerable progress towards meeting the needs of these industrial and business specific applications. However, there are still considerable challenges facing this multidisciplinary field. Drawing from some industry specific applications this talk will cover the trends and challenges facing the researchers and practitioners of this rapidly evolving area. In particular, this talk will outline a set of issues that inhibit or delay the successful completion of an industrial application of KDD. This talk will also point out emerging and significant application areas that demand development of new KDD techniques by the researchers and practitioners. One such area is discovering patterns in temporal data. Another is the evolution of discovery algorithms that respond to changing data forms and streams. Finally, this talk will outline the emerging vertical solutions arena that is driven by business value, which is measured as a progress towards minimizing the gap between the needs of the business user and the accessibility and usability of analytic tools.