DCSO: Dynamic Combination of Detector Scores for Outlier Ensembles
This work addresses a specific problem in outlier detection for data mining applications, offering an incremental improvement over existing combination approaches.
The paper tackles the challenge of selecting and combining outlier scores from multiple base detectors in unsupervised outlier ensembles by proposing DCSO, a framework that dynamically selects top-performing detectors based on local data regions, resulting in consistent performance improvements over static methods on ten benchmark datasets.
Selecting and combining the outlier scores of different base detectors used within outlier ensembles can be quite challenging in the absence of ground truth. In this paper, an unsupervised outlier detector combination framework called DCSO is proposed, demonstrated and assessed for the dynamic selection of most competent base detectors, with an emphasis on data locality. The proposed DCSO framework first defines the local region of a test instance by its k nearest neighbors and then identifies the top-performing base detectors within the local region. Experimental results on ten benchmark datasets demonstrate that DCSO provides consistent performance improvement over existing static combination approaches in mining outlying objects. To facilitate interpretability and reliability of the proposed method, DCSO is analyzed using both theoretical frameworks and visualization techniques, and presented alongside empirical parameter setting instructions that can be used to improve the overall performance.