Concept Drift Detection Based on Anomaly Analysis

Publisher:
Springer International Publishing
Publication Type:
Conference Proceeding
Citation:
Neural Information Processing, 2014, pp. 263 - 270
Issue Date:
2014-11-03
Full metadata record
Files in This Item:
Filename Description Size
ThumbnailConcept Drift Detection Based on Anomaly Analysis.pdfAccepted Manuscript version307.15 kB
Adobe PDF
Abstract In online machine learning, the ability to adapt to new concept quickly is highly desired. In this paper, we propose a novel concept drift detection method, which is called Anomaly Analysis Drift Detection (AADD), to improve the performance of machine learning algorithms under non-stationary environment. The proposed AADD method is based on an anomaly analysis of learner's accuracy associate with the similarity between learners' training domain and test data. This method first identifies whether there are conflicts ...
Please use this identifier to cite or link to this item: