TY - JOUR
T1 - Distributed and weighted extreme learning machine for imbalanced big data learning
AU - Wang, Zhiqiong
AU - Xin, Junchang
AU - Yang, Hongxu
AU - Tian, Shuo
AU - Yu, Ge
AU - Xu, Chenren
AU - Yao, Yudong
N1 - Publisher Copyright:
© 1996-2012 Tsinghua University Press.
PY - 2017/4
Y1 - 2017/4
N2 - The Extreme Learning Machine (ELM) and its variants are effective in many machine learning applications such as Imbalanced Learning (IL) or Big Data (BD) learning. However, they are unable to solve both imbalanced and large-volume data learning problems. This study addresses the IL problem in BD applications. The Distributed and Weighted ELM (DW-ELM) algorithm is proposed, which is based on the MapReduce framework. To confirm the feasibility of parallel computation, first, the fact that matrix multiplication operators are decomposable is illustrated. Then, to further improve the computational efficiency, an Improved DW-ELM algorithm (IDW-ELM) is developed using only one MapReduce job. The successful operations of the proposed DW-ELM and IDW-ELM algorithms are finally validated through experiments.
AB - The Extreme Learning Machine (ELM) and its variants are effective in many machine learning applications such as Imbalanced Learning (IL) or Big Data (BD) learning. However, they are unable to solve both imbalanced and large-volume data learning problems. This study addresses the IL problem in BD applications. The Distributed and Weighted ELM (DW-ELM) algorithm is proposed, which is based on the MapReduce framework. To confirm the feasibility of parallel computation, first, the fact that matrix multiplication operators are decomposable is illustrated. Then, to further improve the computational efficiency, an Improved DW-ELM algorithm (IDW-ELM) is developed using only one MapReduce job. The successful operations of the proposed DW-ELM and IDW-ELM algorithms are finally validated through experiments.
KW - weighted Extreme Learning Machine (ELM); imbalanced big data; MapReduce framework; user-definedcounter
UR - http://www.scopus.com/inward/record.url?scp=85017300293&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85017300293&partnerID=8YFLogxK
U2 - 10.23919/TST.2017.7889638
DO - 10.23919/TST.2017.7889638
M3 - Article
AN - SCOPUS:85017300293
SN - 1007-0214
VL - 22
SP - 160
EP - 173
JO - Tsinghua Science and Technology
JF - Tsinghua Science and Technology
IS - 2
M1 - 7889638
ER -