论文信息 - Capital market surveillance using stream processing

Capital market surveillance using stream processing

Exchanges and regulators need effective tools for surveillance and monitoring for timely detection and prevention of fraudulent activities such as market manipulation, price rigging and insider trading. In this paper we describe how Complex Event Processing (CEP) technology can be used in real time detection of potential fraudulent activity. We introduce the general concepts of CEP and how it can be used in Banking and Financial Markets for real time fraud detection. Finally we demonstrate how event stream processing using a data stream management system (DSMS) is superior to a traditional solution designed using a DBMS. We present results obtained after from using a commercial event stream processing system (IBM's InfoSphere Streams platform) for certain typical low latency fraud detection scenarios.

Debnath Mukherjee | Prateep Misra | Aniruddha Mukherjee | Punit Diwan | Prasun Bhattacharjee

[1] David Luckham,et al. The power of events - an introduction to complex event processing in distributed enterprise systems , 2002, RuleML.

[2] Michael Stonebraker,et al. The 8 requirements of real-time stream processing , 2005, SGMD.

[3] Petra Perner,et al. Data Mining - Concepts and Techniques , 2002, Künstliche Intell..

[4] Philip S. Yu,et al. SPADE: the system s declarative stream processing engine , 2008, SIGMOD Conference.

[5] D. Cumming,et al. Global Market Surveillance , 2007 .

[6] Yoonho Park,et al. Implementing a high-volume, low-latency market data processing system on commodity hardware using IBM middleware , 2009, WHPCF '09.

[7] Adrian Paschke,et al. A reference architecture for Event Processing , 2009, DEBS '09.