A stochastic configuration network based on chaotic sparrow search algorithm

• Published in: Knowledge-based systems Vol. 220; p. 106924
• Main Authors: Zhang, Chenglong, Ding, Shifei
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 23.05.2021; Elsevier Science Ltd
• Subjects: Algorithms; Chaotic; Configurations; Global optimization; Heuristic methods; Logistic mapping; Mutation; Optimization; Parameters; Performance evaluation; Regularization; Search algorithms; Sparrow search algorithm; Stochastic configuration network; Sparrow search algorithm; Stochastic configuration network; Optimization; Chaotic; Logistic mapping
• ISSN: 0950-7051; 1872-7409
• DOI: 10.1016/j.knosys.2021.106924

Stochastic configuration network (SCN), as a novel incremental generation model with supervisory mechanism, has an excellent superiority in solving large-scale data regression and classification problems. However, the accuracy of the SCN is affected by the assignation and selection of some network parameters significantly Sparrow search algorithm (SSA) is a new meta-heuristic algorithm that simulates the foraging and anti-predation behavior of sparrow population. In this paper, a stochastic configuration network based on chaotic sparrow search algorithm is first introduced, termed as CSSA-SCN. Firstly, chaotic sparrow search algorithm (CSSA) is designed which mainly utilizes logistic mapping, self-adaptive hyper-parameters, mutation operator to enhance the global optimization capability of SSA; Secondly, as the performance of SCN is related to regularization parameter r and scale factor λ of weights and biases, then CSSA is employed to give better parameters for SCN automatically; Finally, 13 benchmark functions and several datasets are used to evaluate the performance of CSSA and CSSA-SCN respectively. Experimental results demonstrate the feasibility and validity of CSSA-SCN compared with SCN and other contrast algorithms.