Title: Design and verification of use case generation algorithm based on multiple combination tests

Authors: Genrang Zheng

Addresses: School of Information Engineering, Zhongshan Polytechnic, Zhongshan, 528404, China

Abstract: To solve the problem of low accuracy of interface test cases, an AdaBoost + SVM use case generation algorithm based on multiple combination tests is proposed. Multiple SVM classifiers are homogeneously integrated by using serial AdaBoost integration methods, and they are simulated by using the PICT combination test tools. The results show that the proposed AdaBoost + SVM use case generation algorithm can generate multiple combination interface test cases automatically, and it has a strong interface defect detection capability. Compared with the use case generation algorithm of single SVM classifier and parallel bagging integrated SVM classifier and the current popular use case generation algorithm, the proposed AdaBoost + SVM algorithm performs better in accuracy, precision, recall, F1 value and AUC values, which has certain advantages and can improve the accuracy of interface test.

Keywords: combination test; interface test; use case generation; SVM algorithm; ensemble learning.

DOI: 10.1504/IJBIC.2024.136729

International Journal of Bio-Inspired Computation, 2024 Vol.23 No.2, pp.69 - 79

Received: 19 Aug 2022
Accepted: 18 Aug 2023
Published online: 19 Feb 2024 *

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