File(s) under embargo

Reason: 12 month publisher embargo

11

month(s)

9

day(s)

until file(s) become available

Banyan Tree Growth Optimization and Application

journal contribution
posted on 2023-01-13, 17:18 authored by X Wu, W Zhou, M Fei, Huiyu Zhou

In the era of big data, the number of science and technology service resources has proliferated, and the integration and clustering of science and technology documents become a challenging issue. This paper proposes a novel meta-heuristic algorithm, banyan tree growth optimization (BTGO), for resource clustering of science and technology services. The proposed algorithm is inspired by the growth process of banyan tree, which periodically uses three operators including rooting, multi-trunk, and adjustment to search the solution space globally according to the growth conditions of different stages. To evaluate the performance of BTGO, 29 CEC17 benchmark functions were first utilized to examine its effectiveness. Moreover, a clustering study on UCI datasets is then presented, which compares the suggested algorithm with seven advanced metaheuristic optimization algorithms. The results of numerical experiments and standard datasets demonstrate the effectiveness and efficiency of BTGO. In clustering optimization problems, BTGO can not only finding the optimal solution efficiently, but also improving the clustering accuracy and NMI significantly. Our method was successfully applied to solve the science and technology text clustering problem and validated on the Hainan Science and Technology Service Experimental Platform.

History

Author affiliation

School of Computing and Mathematical Sciences, University of Leicester

Version

AM (Accepted Manuscript)

Published in

Cluster Computing

Publisher

Springer (part of Springer Nature)

issn

1573-7543

Copyright date

2023

Available date

2024-01-07

Language

en

Usage metrics

Categories

Exports