A Genetic Algorithm Based Technique for Outlier Detection with Fast Convergence

Xiaodong Zhu, Ji Zhang, Zewen Hu, Hongzhou Li, Liang Chang, Youwen Zhu, Jerry Chun-Wei Lin, Yongrui Qin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we study the problem of subspace outlier detection in high dimensional data space and propose a new genetic algorithm-based tech- nique to identify outliers embedded in subspaces. The existing technique, mainly using genetic algorithm (GA) to carry out the subspace search, is generally slow due to its expensive fitness evaluation and long solution encoding scheme. In this paper, we propose a novel technique to improve the performance of the exist- ing GA-based outlier detection method using a bit freezing approach to achieve a faster convergence. Through freezing converged bits in the solution encoding strings, this innovative approach can contribute to fast crossover and mutation op- erations and achieve an early stop of the GA that leads to more accurate approxi- mation of fitness function. This research work can contribute to the development of a more efficient search method for detecting subspace outliers. The experimen- tal results demonstrate the improved efficiency of our technique compared with the existing method.
Original languageEnglish
Title of host publicationAdvanced Data Mining and Applications
Subtitle of host publication14th International Conference, ADMA 2018, Nanjing, China, November 16-18, 2018 Proceedings
EditorsGuojun Gan, Bohan Li, Xue Li, Shuliang Wang
Place of PublicationCham
PublisherSpringer Verlag
Pages95-104
Number of pages10
ISBN (Electronic)9783030050900
ISBN (Print)9783030050894
DOIs
Publication statusPublished - 12 Jan 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11323 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Genetic algorithms
Freezing

Cite this

Zhu, X., Zhang, J., Hu, Z., Li, H., Chang, L., Zhu, Y., ... Qin, Y. (2019). A Genetic Algorithm Based Technique for Outlier Detection with Fast Convergence. In G. Gan, B. Li, X. Li, & S. Wang (Eds.), Advanced Data Mining and Applications : 14th International Conference, ADMA 2018, Nanjing, China, November 16-18, 2018 Proceedings (pp. 95-104). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11323 LNAI). Cham: Springer Verlag. https://doi.org/10.1007/978-3-030-05090-0_8
Zhu, Xiaodong ; Zhang, Ji ; Hu, Zewen ; Li, Hongzhou ; Chang, Liang ; Zhu, Youwen ; Lin, Jerry Chun-Wei ; Qin, Yongrui. / A Genetic Algorithm Based Technique for Outlier Detection with Fast Convergence. Advanced Data Mining and Applications : 14th International Conference, ADMA 2018, Nanjing, China, November 16-18, 2018 Proceedings. editor / Guojun Gan ; Bohan Li ; Xue Li ; Shuliang Wang. Cham : Springer Verlag, 2019. pp. 95-104 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b2cb0a6d703346799fdf0db521345966,
title = "A Genetic Algorithm Based Technique for Outlier Detection with Fast Convergence",
abstract = "In this paper, we study the problem of subspace outlier detection in high dimensional data space and propose a new genetic algorithm-based tech- nique to identify outliers embedded in subspaces. The existing technique, mainly using genetic algorithm (GA) to carry out the subspace search, is generally slow due to its expensive fitness evaluation and long solution encoding scheme. In this paper, we propose a novel technique to improve the performance of the exist- ing GA-based outlier detection method using a bit freezing approach to achieve a faster convergence. Through freezing converged bits in the solution encoding strings, this innovative approach can contribute to fast crossover and mutation op- erations and achieve an early stop of the GA that leads to more accurate approxi- mation of fitness function. This research work can contribute to the development of a more efficient search method for detecting subspace outliers. The experimen- tal results demonstrate the improved efficiency of our technique compared with the existing method.",
author = "Xiaodong Zhu and Ji Zhang and Zewen Hu and Hongzhou Li and Liang Chang and Youwen Zhu and Lin, {Jerry Chun-Wei} and Yongrui Qin",
year = "2019",
month = "1",
day = "12",
doi = "10.1007/978-3-030-05090-0_8",
language = "English",
isbn = "9783030050894",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "95--104",
editor = "Guojun Gan and Bohan Li and Xue Li and Shuliang Wang",
booktitle = "Advanced Data Mining and Applications",

}

Zhu, X, Zhang, J, Hu, Z, Li, H, Chang, L, Zhu, Y, Lin, JC-W & Qin, Y 2019, A Genetic Algorithm Based Technique for Outlier Detection with Fast Convergence. in G Gan, B Li, X Li & S Wang (eds), Advanced Data Mining and Applications : 14th International Conference, ADMA 2018, Nanjing, China, November 16-18, 2018 Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11323 LNAI, Springer Verlag, Cham, pp. 95-104. https://doi.org/10.1007/978-3-030-05090-0_8

A Genetic Algorithm Based Technique for Outlier Detection with Fast Convergence. / Zhu, Xiaodong; Zhang, Ji; Hu, Zewen; Li, Hongzhou; Chang, Liang; Zhu, Youwen; Lin, Jerry Chun-Wei; Qin, Yongrui.

Advanced Data Mining and Applications : 14th International Conference, ADMA 2018, Nanjing, China, November 16-18, 2018 Proceedings. ed. / Guojun Gan; Bohan Li; Xue Li; Shuliang Wang. Cham : Springer Verlag, 2019. p. 95-104 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11323 LNAI).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A Genetic Algorithm Based Technique for Outlier Detection with Fast Convergence

AU - Zhu, Xiaodong

AU - Zhang, Ji

AU - Hu, Zewen

AU - Li, Hongzhou

AU - Chang, Liang

AU - Zhu, Youwen

AU - Lin, Jerry Chun-Wei

AU - Qin, Yongrui

PY - 2019/1/12

Y1 - 2019/1/12

N2 - In this paper, we study the problem of subspace outlier detection in high dimensional data space and propose a new genetic algorithm-based tech- nique to identify outliers embedded in subspaces. The existing technique, mainly using genetic algorithm (GA) to carry out the subspace search, is generally slow due to its expensive fitness evaluation and long solution encoding scheme. In this paper, we propose a novel technique to improve the performance of the exist- ing GA-based outlier detection method using a bit freezing approach to achieve a faster convergence. Through freezing converged bits in the solution encoding strings, this innovative approach can contribute to fast crossover and mutation op- erations and achieve an early stop of the GA that leads to more accurate approxi- mation of fitness function. This research work can contribute to the development of a more efficient search method for detecting subspace outliers. The experimen- tal results demonstrate the improved efficiency of our technique compared with the existing method.

AB - In this paper, we study the problem of subspace outlier detection in high dimensional data space and propose a new genetic algorithm-based tech- nique to identify outliers embedded in subspaces. The existing technique, mainly using genetic algorithm (GA) to carry out the subspace search, is generally slow due to its expensive fitness evaluation and long solution encoding scheme. In this paper, we propose a novel technique to improve the performance of the exist- ing GA-based outlier detection method using a bit freezing approach to achieve a faster convergence. Through freezing converged bits in the solution encoding strings, this innovative approach can contribute to fast crossover and mutation op- erations and achieve an early stop of the GA that leads to more accurate approxi- mation of fitness function. This research work can contribute to the development of a more efficient search method for detecting subspace outliers. The experimen- tal results demonstrate the improved efficiency of our technique compared with the existing method.

UR - http://www.scopus.com/inward/record.url?scp=85059748692&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-05090-0_8

DO - 10.1007/978-3-030-05090-0_8

M3 - Conference contribution

SN - 9783030050894

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 95

EP - 104

BT - Advanced Data Mining and Applications

A2 - Gan, Guojun

A2 - Li, Bohan

A2 - Li, Xue

A2 - Wang, Shuliang

PB - Springer Verlag

CY - Cham

ER -

Zhu X, Zhang J, Hu Z, Li H, Chang L, Zhu Y et al. A Genetic Algorithm Based Technique for Outlier Detection with Fast Convergence. In Gan G, Li B, Li X, Wang S, editors, Advanced Data Mining and Applications : 14th International Conference, ADMA 2018, Nanjing, China, November 16-18, 2018 Proceedings. Cham: Springer Verlag. 2019. p. 95-104. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-05090-0_8