W. Nick Street and YongSeog Kim. A Streaming Ensemble Algorithm (SEA) for Large-Scale Classification. In Proceedings of the 7th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '01), pages 377--382, 2001.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....tests are given in Table 6 and 5. 7. DISCUSSION AND RELATED WORK Data stream processing has recently become a very important research domain. Much work has been done on modeling [1] querying [2, 13, 16] and mining data streams, for instance, several papers have been published on classification [7, 19, 23], regression analysis [4] and clustering [17] Traditional data mining algorithms are challenged by two characteristic features of data streams: the infinite data flow and the drifting concepts. As methods that require multiple scans of the datasets [21] can not handle infinite data flows, ....

W. Nick Street and YongSeog Kim. A streaming ensemble algorithm (SEA) for large-scale classification. In Int'l Conf. on Knowledge Discovery and Data Mining (SIGKDD), 2001.

....de ned by a weighted average of the outputs of these trees. Breiman proposed an arcing method as the means to learn from large data sets and stream data [3; 5] It is also based on adaptive re sampling. However, it uses unweighted average to build the ensemble classi er. Recently Street and Kim [28] proposed Streaming Ensemble Algorithm (SEA) that learns an ensemble of decision trees for large scale classi cation. SEA maintains a xed number of classi ers. Once the ensemble becomes full, the k th classi er Ck is added only if it outperforms any previous C i in the ensemble. In that case, C ....

N. W. Street and Y. Kim. A streaming ensemble algorithm (sea) for large-scale classi caiton. In Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, pages 377-382, 2001.

....and or require too much memory. Apart from specialized algorithms for particular classi cation models, several generic remedies for the above problems have been proposed in the literature. They can be broadly classi ed into subsampling strategies [8, 13] and learning using committee machines [11, 3, 4, 12, 14]. Of these two strategies, the latter one appears to be particularly promising because (a) it does not require any data to be discarded when building the classi er, and (b) it allows for incremental learning because the model can be updated when a new chunk of data arrives. The basic idea of ....

....method retains a xed size window of weak classi ers that contains the k most recently built classi ers. This makes the method applicable to large datasets in terms of memory and time requirements. However, it remains unclear how an appropriate value for k can be determined. Street and Kim [14] propose a variant of bagging for incremental learning based on data chunks that maintains a xed size committee. In each iteration it attempts to identify a committee member that should be replaced by the model built from the most recent chunk of data. Because the algorithm is based on bagging ....

W. Nick Street and YongSeog Kim. A streaming ensemble algorithm (SEA) for large-scale classi cation. In 7th ACM SIGKDD Int. Conf. on Knowledge Discovery in Databases and Data Mining, pages 377-382, 2001.

No context found.

W.N. Street and Y. Kim. A streaming ensemble algorithm (SEA) for large-scale classification. In Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-01), pages 377--382, 2001.

No context found.

W. Nick Street and YongSeog Kim. A Streaming Ensemble Algorithm (SEA) for Large-Scale Classification. In Proceedings of the 7th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '01), pages 377--382, 2001.

No context found.

W. Nick Street and YongSeog Kim. A streaming ensemble algorithm (sea) for large-scale classification. In International Conference on Knowledge Discovery and Data Mining (SIGKDD), 2001.

No context found.

W. Street and Y. Kim. A streaming ensemble algorithm(sea) for large-scale classification. In ACM SIGKDD, 2001.

No context found.

W. Street and Y. Kim. A streaming ensemble algorithm (SEA) for large-scale classification. In Proceedings of the 7th ACM International Conference on Knowledge Discovery and Data Mining, pages 377--382, ACM Press, New York, NY, 2001.

No context found.

W. Street and Y. Kim. A streaming ensemble algorithm(sea) for large-scale classi cation. In ACM SIGKDD, 2001.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC