Extreme Learning Machine with Multi-Agent System for Regression
Chong Tak Yaw1, Keem Siah Yap2, Shen Yuong Wong3, Chin Hooi Tan4
1Chong Tak Yaw, Department of Electrical and Electronics Engineering, Universiti Tenaga Nasional, Malaysia.
2Keem Siah Yap, Department of Electrical and Electronics Engineering, Universiti Tenaga Nasional, Malaysia.
3Shen Yuong Wong, Department of Electrical and Electronics Engineering, Xiamen University Malaysia, Malaysia.
4Chin Hooi Tan, Business Innovation Incubator, Tenaga Nasional Berhad, Malaysia.
Manuscript received on November 22, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP: 4149-4153 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B4922129219/2019©BEIESP | DOI: 10.35940/ijeat.B4922.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: From the point of learning speed as well as generalization, Extreme Learning Machine(ELM) is widely known as an effective learning algorithm than the conventional learning methods. Basically, hidden neurons are not required in neuron alike, instead, weight is the parameter that would need to learn about the link in between output and hidden layers. The creation of an output is to integrate each independent of several ELMs. The precise approach is included in a Multi-Agent System. The novelty of ELM-MAS (extreme learning machine based multi-agent system) is put forward in the paper for solving data regression problems. The ELMs consist of two layers which are the parent agent layer and individual agent layer. The effectiveness of the ELM-MAS model is proved with some activation functions employing benchmark datasets (abalone, strike and space-ga) and real world application (Nox emission). The outcomes indicate that the proposed model is capable to attain improved results than other approaches.
Keywords: Extreme Learning Machine (ELM); Multi Agent System (MAS); Data Regression; NOx Emission of Power Plant.