Concept Drift Detection based on Anomaly Analysis

Liu, A; Zhang, G; Lu, J

Concept Drift Detection based on Anomaly Analysis

Liu, A Zhang, G Lu, J

Permalink

Publication Type:: Conference
Citation:: 2014
Issue Date:: 2014-12-01

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Accepted Manuscript versionAdobe PDF (462.49 kB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Liu, A
dc.contributor.author	Zhang, G
dc.contributor.author	Lu, J
dc.contributor.editor	Loo, CK
dc.contributor.editor	Keem Siah, Y
dc.contributor.editor	Wong, KKW
dc.contributor.editor	Beng Jin, AT
dc.contributor.editor	Huang, K
dc.date	2014-11-03
dc.date.accessioned	2015-05-19T10:56:47Z
dc.date.available	2015-05-19T10:56:47Z
dc.date.issued	2014-12-01
dc.identifier.citation	2014
dc.identifier.uri	http://hdl.handle.net/10453/35694
dc.description.abstract	Abstract. In online machine learning, the ability to adapt to new concept quick-ly is highly desired. In this paper, we propose a novel concept drift detection method, which is called Anomaly Analysis Drift Detection (AADD), to im-prove the performance of machine learning algorithms under non-stationary en-vironment. The proposed AADD method is based on an anomaly analysis of learner’s accuracy associate with the similarity between learners’ training do-main and test data. This method first identifies whether there are conflicts be-tween current concept and new coming data. Then the learner will incremental-ly learn the non-conflict data, which will not decrease the accuracy of the learn-er on previous trained data, for concept extension. Otherwise, a new learner will be created based on the new data. Experiments illustrate that this AADD meth-od can detect new concept quickly and learn extensional drift incrementally.
dc.relation.ispartof	International Conference on Neural Information Processing
dc.relation.isreplacedby	10453/121349
dc.relation.isreplacedby	http://hdl.handle.net/10453/121349
dc.title	Concept Drift Detection based on Anomaly Analysis
dc.type	Conference
dc.location.activity	Kuching, Sarawak
pubs.embargo.period	Not known
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Software
pubs.organisational-group	/University of Technology Sydney/Strength - Quantum Computation and Intelligent Systems
pubs.organisational-group	/University of Technology Sydney/Students
utslib.copyright.status	open_access
pubs.declined	1970-01-01T00:00:00.0+1000
pubs.consider-herdc	false
pubs.merge-to	10453/121349
pubs.merge-to	http://hdl.handle.net/10453/121349
pubs.deleted	1970-01-01T00:00:00.0+1000

Abstract:

Abstract. In online machine learning, the ability to adapt to new concept quick-ly is highly desired. In this paper, we propose a novel concept drift detection method, which is called Anomaly Analysis Drift Detection (AADD), to im-prove the performance of machine learning algorithms under non-stationary en-vironment. The proposed AADD method is based on an anomaly analysis of learner’s accuracy associate with the similarity between learners’ training do-main and test data. This method first identifies whether there are conflicts be-tween current concept and new coming data. Then the learner will incremental-ly learn the non-conflict data, which will not decrease the accuracy of the learn-er on previous trained data, for concept extension. Otherwise, a new learner will be created based on the new data. Experiments illustrate that this AADD meth-od can detect new concept quickly and learn extensional drift incrementally.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/35694