Random forest based approach for concept drift handling

Aleksei V. Zhukov; Denis N. Sidorov; Aoife M. Foley

doi:10.1007/978-3-319-52920-2_7

Random forest based approach for concept drift handling

Aleksei V. Zhukov^*
, Denis N. Sidorov
, Aoife M. Foley

^*Corresponding author for this work

Melent'ev Institute of Power Engineering Systems
Irkutsk State University
Queen's University Belfast

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

49 Scopus citations

Abstract

Concept drift has potential in smart grid analysis because the socio-economic behaviour of consumers is not governed by the laws of physics. Likewise there are also applications in wind power forecasting. In this paper we present decision tree ensemble classification method based on the Random Forest algorithm for concept drift. The weighted majority voting ensemble aggregation rule is employed based on the ideas of Accuracy Weighted Ensemble (AWE) method. Base learner weight in our case is computed for each sample evaluation using base learners accuracy and intrinsic proximity measure of Random Forest. Our algorithm exploits ensemble pruning as a forgetting strategy. We present results of empirical comparison of our method and other state-of-the-art concept-drfit classifiers.

Original language	English
Title of host publication	Analysis of Images, Social Networks and Texts - 5th International Conference, AIST 2016, Revised Selected Papers
Editors	Natalia Loukachevitch, Alexander Panchenko, Konstantin Vorontsov, Valeri G. Labunets, Andrey V. Savchenko, Dmitry I. Ignatov, Sergey I. Nikolenko, Mikhail Yu. Khachay
Publisher	Springer Verlag
Pages	69-77
Number of pages	9
ISBN (Print)	9783319529196
DOIs	https://doi.org/10.1007/978-3-319-52920-2_7
State	Published - 2017
Externally published	Yes
Event	5th International Conference on Analysis of Images, Social Networks and Texts, AIST 2016 - Yekaterinburg, Russian Federation Duration: 7 Apr 2016 → 9 Apr 2016

Publication series

Name	Communications in Computer and Information Science
Volume	661
ISSN (Print)	1865-0929

Conference

Conference	5th International Conference on Analysis of Images, Social Networks and Texts, AIST 2016
Country/Territory	Russian Federation
City	Yekaterinburg
Period	7/04/16 → 9/04/16

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Classification
Concept drift
Decision tree
Ensemble learning
Machine learning
Random forest

Access to Document

10.1007/978-3-319-52920-2_7

Cite this

Zhukov, A. V., Sidorov, D. N., & Foley, A. M. (2017). Random forest based approach for concept drift handling. In N. Loukachevitch, A. Panchenko, K. Vorontsov, V. G. Labunets, A. V. Savchenko, D. I. Ignatov, S. I. Nikolenko, & M. Y. Khachay (Eds.), Analysis of Images, Social Networks and Texts - 5th International Conference, AIST 2016, Revised Selected Papers (pp. 69-77). (Communications in Computer and Information Science; Vol. 661). Springer Verlag. https://doi.org/10.1007/978-3-319-52920-2_7

Zhukov, Aleksei V. ; Sidorov, Denis N. ; Foley, Aoife M. / Random forest based approach for concept drift handling. Analysis of Images, Social Networks and Texts - 5th International Conference, AIST 2016, Revised Selected Papers. editor / Natalia Loukachevitch ; Alexander Panchenko ; Konstantin Vorontsov ; Valeri G. Labunets ; Andrey V. Savchenko ; Dmitry I. Ignatov ; Sergey I. Nikolenko ; Mikhail Yu. Khachay. Springer Verlag, 2017. pp. 69-77 (Communications in Computer and Information Science).

@inproceedings{738c9c1e22b0433ebb40ecd237bdb4fd,

title = "Random forest based approach for concept drift handling",

abstract = "Concept drift has potential in smart grid analysis because the socio-economic behaviour of consumers is not governed by the laws of physics. Likewise there are also applications in wind power forecasting. In this paper we present decision tree ensemble classification method based on the Random Forest algorithm for concept drift. The weighted majority voting ensemble aggregation rule is employed based on the ideas of Accuracy Weighted Ensemble (AWE) method. Base learner weight in our case is computed for each sample evaluation using base learners accuracy and intrinsic proximity measure of Random Forest. Our algorithm exploits ensemble pruning as a forgetting strategy. We present results of empirical comparison of our method and other state-of-the-art concept-drfit classifiers.",

keywords = "Classification, Concept drift, Decision tree, Ensemble learning, Machine learning, Random forest",

author = "Zhukov, \{Aleksei V.\} and Sidorov, \{Denis N.\} and Foley, \{Aoife M.\}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 5th International Conference on Analysis of Images, Social Networks and Texts, AIST 2016 ; Conference date: 07-04-2016 Through 09-04-2016",

year = "2017",

doi = "10.1007/978-3-319-52920-2\_7",

language = "英语",

isbn = "9783319529196",

series = "Communications in Computer and Information Science",

publisher = "Springer Verlag",

pages = "69--77",

editor = "Natalia Loukachevitch and Alexander Panchenko and Konstantin Vorontsov and Labunets, \{Valeri G.\} and Savchenko, \{Andrey V.\} and Ignatov, \{Dmitry I.\} and Nikolenko, \{Sergey I.\} and Khachay, \{Mikhail Yu.\}",

booktitle = "Analysis of Images, Social Networks and Texts - 5th International Conference, AIST 2016, Revised Selected Papers",

address = "德国",

}

Zhukov, AV, Sidorov, DN & Foley, AM 2017, Random forest based approach for concept drift handling. in N Loukachevitch, A Panchenko, K Vorontsov, VG Labunets, AV Savchenko, DI Ignatov, SI Nikolenko & MY Khachay (eds), Analysis of Images, Social Networks and Texts - 5th International Conference, AIST 2016, Revised Selected Papers. Communications in Computer and Information Science, vol. 661, Springer Verlag, pp. 69-77, 5th International Conference on Analysis of Images, Social Networks and Texts, AIST 2016, Yekaterinburg, Russian Federation, 7/04/16. https://doi.org/10.1007/978-3-319-52920-2_7

Random forest based approach for concept drift handling. / Zhukov, Aleksei V.; Sidorov, Denis N.; Foley, Aoife M.
Analysis of Images, Social Networks and Texts - 5th International Conference, AIST 2016, Revised Selected Papers. ed. / Natalia Loukachevitch; Alexander Panchenko; Konstantin Vorontsov; Valeri G. Labunets; Andrey V. Savchenko; Dmitry I. Ignatov; Sergey I. Nikolenko; Mikhail Yu. Khachay. Springer Verlag, 2017. p. 69-77 (Communications in Computer and Information Science; Vol. 661).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Random forest based approach for concept drift handling

AU - Zhukov, Aleksei V.

AU - Sidorov, Denis N.

AU - Foley, Aoife M.

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - Concept drift has potential in smart grid analysis because the socio-economic behaviour of consumers is not governed by the laws of physics. Likewise there are also applications in wind power forecasting. In this paper we present decision tree ensemble classification method based on the Random Forest algorithm for concept drift. The weighted majority voting ensemble aggregation rule is employed based on the ideas of Accuracy Weighted Ensemble (AWE) method. Base learner weight in our case is computed for each sample evaluation using base learners accuracy and intrinsic proximity measure of Random Forest. Our algorithm exploits ensemble pruning as a forgetting strategy. We present results of empirical comparison of our method and other state-of-the-art concept-drfit classifiers.

AB - Concept drift has potential in smart grid analysis because the socio-economic behaviour of consumers is not governed by the laws of physics. Likewise there are also applications in wind power forecasting. In this paper we present decision tree ensemble classification method based on the Random Forest algorithm for concept drift. The weighted majority voting ensemble aggregation rule is employed based on the ideas of Accuracy Weighted Ensemble (AWE) method. Base learner weight in our case is computed for each sample evaluation using base learners accuracy and intrinsic proximity measure of Random Forest. Our algorithm exploits ensemble pruning as a forgetting strategy. We present results of empirical comparison of our method and other state-of-the-art concept-drfit classifiers.

KW - Classification

KW - Concept drift

KW - Decision tree

KW - Ensemble learning

KW - Machine learning

KW - Random forest

UR - https://www.scopus.com/pages/publications/85014174917

U2 - 10.1007/978-3-319-52920-2_7

DO - 10.1007/978-3-319-52920-2_7

M3 - 会议稿件

AN - SCOPUS:85014174917

SN - 9783319529196

T3 - Communications in Computer and Information Science

SP - 69

EP - 77

BT - Analysis of Images, Social Networks and Texts - 5th International Conference, AIST 2016, Revised Selected Papers

A2 - Loukachevitch, Natalia

A2 - Panchenko, Alexander

A2 - Vorontsov, Konstantin

A2 - Labunets, Valeri G.

A2 - Savchenko, Andrey V.

A2 - Ignatov, Dmitry I.

A2 - Nikolenko, Sergey I.

A2 - Khachay, Mikhail Yu.

PB - Springer Verlag

T2 - 5th International Conference on Analysis of Images, Social Networks and Texts, AIST 2016

Y2 - 7 April 2016 through 9 April 2016

ER -

Zhukov AV, Sidorov DN, Foley AM. Random forest based approach for concept drift handling. In Loukachevitch N, Panchenko A, Vorontsov K, Labunets VG, Savchenko AV, Ignatov DI, Nikolenko SI, Khachay MY, editors, Analysis of Images, Social Networks and Texts - 5th International Conference, AIST 2016, Revised Selected Papers. Springer Verlag. 2017. p. 69-77. (Communications in Computer and Information Science). doi: 10.1007/978-3-319-52920-2_7

Random forest based approach for concept drift handling

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this