Accuracy weighted diversity-based online boosting

被引：10

作者：

Baidari, Ishwar ^{[1
]}

Honnikoll, Nagaraj ^{[1
]}

机构：

[1] Karnatak Univ, Dept Comp Sci, Dharwad 580003, Karnataka, India

来源：

EXPERT SYSTEMS WITH APPLICATIONS | 2020年 / 160卷

关键词：

Data stream; Concept drift; Online boosting; Diversity; DRIFT DETECTION; ENSEMBLE; CLASSIFIERS; MAJORITY;

D O I：

10.1016/j.eswa.2020.113723

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Target distributional change occurring in a data stream known as concept drift, causes a challenging task for an online learning method, as the accuracy of an online learning method may decrease due to these changes. In this paper, the Accuracy Weighted Diversity-based Online Boosting (AWDOB) method has been proposed, which is based on Adaptable Diversity-based Online Boosting (ADOB) and, other modifications. More precisely, AWDOB uses the proposed accuracy weighting scheme which is based on previous expert's results of the sums of correctly classified and incorrectly classified instances to calculate the weight of current expert, which improved the overall accuracy of the AWDOB. Experiments were conducted to compare the accuracy results of AWDOB against other methods using ten real-world datasets and thirty-two artificial datasets. Artificial datasets were generated by the four artificial data generators which included gradual and abrupt concept drifts within them. Experimental results suggest that AWDOB beats the accuracy results of other tested methods. (c) 2020 Elsevier Ltd. All rights reserved.

引用

页数：16

共 44 条

[21]

Garrido Teixeira de Carvalho Santos Silas, 2014, Machine Learning and Knowledge Discovery in Databases. European Conference, ECML PKDD 2014. Proceedings: LNCS 8726, P179, DOI 10.1007/978-3-662-44845-8_12

[22]

Hulten G., 2001, KDD-2001. Proceedings of the Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, P97, DOI 10.1145/502512.502529

[23]

Jedrzejowicz Joanna, 2020, Intelligent Decision Technologies 2019. Proceedings of the 11th KES International Conference on Intelligent Decision Technologies (KES-IDT 2019). Smart Innovation, Systems and Technologies (SIST 142), P203, DOI 10.1007/978-981-13-8311-3_18

[24]

Khamassi I., 2018, EVOL SYST-GER, V9, P1, DOI DOI 10.1007/s12530-016-9168-2

[25] STATLOG - COMPARISON OF CLASSIFICATION ALGORITHMS ON LARGE REAL-WORLD PROBLEMS [J].

KING, RD ;

FENG, C ;

SUTHERLAND, A .

APPLIED ARTIFICIAL INTELLIGENCE, 1995, 9 (03) :289-333

[26]

Kolter JZ, 2007, J MACH LEARN RES, V8, P2755

[27] Online ensemble learning with abstaining classifiers for drifting and noisy data streams [J].

Krawczyk, Bartosz ;

Cano, Alberto .

APPLIED SOFT COMPUTING, 2018, 68 :677-692

[28] Concept Drift Detection and Adaption in Big Imbalance Industrial IoT Data Using an Ensemble Learning Method of Offline Classifiers [J].

Lin, Chun-Cheng ;

Deng, Der-Jiunn ;

Kuo, Chin-Hung ;

Chen, Linnan .

IEEE ACCESS, 2019, 7 :56198-56207

[29] THE WEIGHTED MAJORITY ALGORITHM [J].

LITTLESTONE, N ;

WARMUTH, MK .

INFORMATION AND COMPUTATION, 1994, 108 (02) :212-261

[30] Adaptive learning on mobile network traffic data [J].

Liu, Zhen ;

Japkowicz, Nathalie ;

Wang, Ruoyu ;

Tang, Deyu .

CONNECTION SCIENCE, 2019, 31 (02) :185-214

← 1 2 3 4 5 →