Raphael Ducasse
ABSTRACT
Multi-concept identification in high volume multimedia streams is critical for a number of applications, including large-scale multimedia analysis, processing, and retrieval. Content of interest is filtered using a collection of binary classifiers that are deployed on distributed resource-constrained infrastructure. In this paper, we design distributed algorithms for determining the optimal topology of single concept detectors (classifiers) to identify the multiple concepts of interest. These algorithms dynamically order individual classifiers into chain topologies to tradeoff accuracy against processing delay, based on underlying data characteristics, system resource constraints as well as the performance and complexity characteristics of each classifier
- B. Babcock, S. Babu, R. Motwani, and M. Datar. Chain: operator scheduling for memory minimization in data stream systems. In ACM SIGMOD, 2003. Google Scholar
Digital Library
- S. Babu, R. Motwani, K. Munagala, I. Nishizawa, and J. Widom. Adaptive ordering of pipelined stream filters. ACM SIGMOD, 2004. Google ScholarDigital Library
- R. Ducasse, D. S. Turaga, and M. van der Schaar. Adaptive topologic optimization for large-scale stream mining. IEEE Journal on Selected Topics in Signal Processing, under review.Google Scholar
- F. Fu, D. S. Turaga, O. Verscheure, M. van der Schaar, and L. Amini. Configuring competing classifier chains in distributed stream mining systems. IEEE Journal on Selected Topics in Signal Processing, 2007.Google Scholar
- R. Lienhart, L. Liang, and A. Kuranov. A detector tree of boosted classifiers for real-time objects detection and tracking. In IEEE ICME, 2003. Google ScholarDigital Library
- H. Park, D. S. Turaga, O. Verscheure, and M. van der Schaar. Foresighted tree configuring games in resource constrained distributed stream mining systems. IEEE ICASSP, 2009. Google ScholarDigital Library
- L. Saul and M. I. Jordan. Learning in boltzman trees. Neural Computation, 1994. Google ScholarDigital Library
- R. S. Sutton and A. G. Barto. Reinforcement learning: An introduction. The MIT Press, Cambridge, MA,, 1998. Google ScholarDigital Library
- V. Vazirani. Approximation algorithms. Springer Verlag. Google ScholarDigital Library
Index Terms
- Topology selection for stream mining systems
Recommendations
Configuring topologies of distributed semantic concept classifiers for continuous multimedia stream processing
MM '08: Proceedings of the 16th ACM international conference on MultimediaReal-time multimedia semantic concept detection requires instant identification of a set of concepts in streaming video or images. However, the potentially high data volumes of multimedia content, and high complexity associated with individual concept ...
Data Stream Mining: Challenges and Techniques
ICTAI '10: Proceedings of the 2010 22nd IEEE International Conference on Tools with Artificial Intelligence - Volume 02Data streams are continuous flows of data. Examples of data streams include network traffic, sensor data, call center records and so on. Their sheer volume and speed pose a great challenge for the data mining community to mine them. Data streams ...
Stream mining: a novel architecture for ensemble-based classification
Mining data streams has become an important and challenging task for a wide range of applications. In these scenarios, data tend to arrive in multiple, rapid and time-varying streams, thus constraining data mining algorithms to look at data only once. ...
Comments