Association for Computing Machinery
Outlier detection and ensemble learning are well established research directions in data mining yet the application of ensemble techniques to outlier detection has been rarely studied. Here, the authors propose and study sub-sampling as a technique to induce diversity among individual outlier detectors. They show analytically and experimentally that an outlier detector based on a subsample per se, besides inducing diversity, can, under certain conditions, already improve upon the results of the same outlier detector on the complete dataset. Building an ensemble on top of several subsamples is further improving the results.