Title: An anomaly feature mining method for software test data based on bat algorithm

Authors: Yirong Guo; Jieli Chen

Addresses: School of Information Engineering, Longdong University, Xifeng, Qingyang, Gansu, China ' College of Mathematical and Information Engineering, Longyan University, Xinluo, Longyan, China

Abstract: Traditional data anomaly feature mining methods usually have low accuracy and efficiency, so a new method based on bat algorithm is proposed to mine test data anomaly features. First of all, according to the distribution sequence of test data, combined with the data correlation analysis results into data collection. Secondly, the spatial distribution function of software test data features is constructed to complete the analysis of data feature correlation. Finally, the optimal objective function of software test data anomaly feature mining is constructed, and the bat algorithm is used to solve the objective function to obtain data anomaly features. The results show that this research method has a high effect on improving the accuracy and efficiency, and the accuracy of anomaly feature mining is always above 90%.

Keywords: bat algorithm; software test data; anomaly feature mining; objective function.

DOI: 10.1504/IJDMB.2022.130338

International Journal of Data Mining and Bioinformatics, 2022 Vol.27 No.1/2/3, pp.58 - 72

Received: 30 Sep 2022
Accepted: 15 Dec 2022
Published online: 17 Apr 2023 *

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