Title: GA-IPSO-BSVM based sustainable development of ecological economic logistics data forecasting model

Authors: Yan Sun

Addresses: School of Business, Xi'an International University, Xi'an 710077, Shaanxi Province, China

Abstract: Considering the increasing accuracy of machine learning in detection and recognition tasks, this paper proposes a classification model, which consists of an improved genetic algorithm (GA), the improved particle swarm optimisation (IPSO) and balanced support vector machine (SVM) (namely GA-IPSO-BSVM), to increase the accuracy and to optimise the convergence of logistics data performance classification. Firstly, this paper embeds the elimination mechanism of GA to my model at the early stage of iterations to delete a lot of particles with low speed. Then, in the middle of the iteration, this paper improves the topology structure of particle relationship in PSO to avoid the algorithm trapping into local optimal solution. At the late stage of the iteration, this paper combines the excellent particles in all regions into the excellent particle population and iterates the population to obtain the global optimal solution.

Keywords: logistics data; ecological-economic belt; GA; particle swarm optimisation; PSO; support vector machine; SVM.

DOI: 10.1504/IJESD.2025.142932

International Journal of Environment and Sustainable Development, 2025 Vol.24 No.1, pp.22 - 34

Received: 18 May 2023
Accepted: 31 Jul 2023
Published online: 02 Dec 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article