Ensemble online sequential extreme learning machine for large data set classification

Abstract

Online sequential extreme learning machine (OS-ELM) proposed by Liang et al. employ sequential learning strategy to learn the target concept from the data. Compared with the original ELM, OS-ELM can learn data one-by-one or chunk-by-chunk with fixed or varying chunk size with almost same performance as ELM. While compared with other state-ofthe-art sequential algorithms such as SGBP, RAN and GAP-RBF, OS-ELM has faster learning speed and better generalization ability. However, similar to ELM, OS-ELM also has instability in different trials of simulations. In addition, for large data sets, OS-ELM will not halt when there are training samples not be learned, this phenomenon results in long learning time. In order to deal with the problems, this paper proposes an algorithm named E-OS-ELM for integrating OS-ELM to classify large data sets. The experimental results show that the proposed method is effective and efficient; it can effectively overcome the drawbacks mentioned above.

Concept Drift Evolution In Machine Learning Approaches: A Systematic Literature Review

“…(Budiman et al, 2016;Iwashita et al, 2019;Jameel et al, 2020a, b, c;Budiman et al, 2017;Zang et al, 2014;Zliobaite et al, 2014;Kuncheva, 2004;Ghorbani et al, 2017;Gupta and Dhawan, 2019;Jensen et al, 2019;Nishida et al, 2008;Harel et al, 2014;Dyer and Polikar, 2012;Khamassi et al, 2019;Saurav et al, 2018;Dongre and Malik, 2014;Dariusz, 2010;Sayed et al, 2018;Wadewale and Desai, 2015;Brzezinski and Stefanowski, 2014a;Hoens et al, 2012;Jagadeesh Chandra Bose et al, 2011;Huang et al, 2013;Minku et al, 2010;Tsymbal, 2004;Gomes et al, 2011;Hoens et al, 2011;Webb et al, 2016Webb et al, , 2018 To investigate the existing CD handling approaches and determine their effectiveness and shortcomings for current and future trends. (Jameel et al, 2020(Jameel et al, , 2020Uddin et al, 2019, September;Liu and Wang, 2010;van Schaik and Tapson, 2015;Budiman et al, 2016;Huang et al, 2012;Zhai et al, 2014;Xu and Wang, 2016;2017;Khamassi et al, 2019;…”

“…Few recent studies concentrated on adaptive learning techniques using ELM based Single classifiers (van Schaik and Tapson, 2015;Huang et al, 2012; and Ensemble classifier for CD adaptation (Zhai et al, 2014;Xu and Wang, 2016;2017). However, all these solutions lie in the semi-adaptive category (does not implement fully autonomous learning behavior).…”

“…In this approach, the number of neurons in hidden layers and the selection of the activation layer are dynamic, enhancing the performance of the model. At the same time, this approach handles stream data for gradual drift scenario only (Zhai et al, 2014).…”

“…The adaptive classifiers can handle this issue in a better way. Few recent studies concentrated on adaptive learning techniques using ELM based single classifiers (van Schaik and Tapson, 2015;Budiman et al, 2016;Huang et al, 2012) and ensemble classifiers for CD adaptation (Zhai et al, 2014;Wang, 2016, 2017). For example, Incremental Data Stream ELM used an incremental approach to train the classifier.…”

Short-Term Load Forecasting Based on the Analysis of User Electricity Behavior

“…In the traditional extreme learning machine algorithm, when new data are obtained, the historical data will be repeatedly trained together with new data, which requires a lot of time. OS-ELM effectively avoids the repeated training of data and greatly improves the learning efficiency by using a partitioned matrix method [34][35][36].…”