Abdullah Mohamed Abdelmoumen
Demonstrator
SOFTWARE TESTING AUTOMATION USING MACHINE LEARNING TECHNIQUES
International Journal of Advanced Computational Engineering and Networking • 2024
Publication Information
Authors
M Abdul-salam; M Abdul-fattah; Abdullah Mohamed
Keywords
Software testing automation, classification algorithms, deep learning, particle swarm optimization
Journal
International Journal of Advanced Computational Engineering and Networking
Publisher
Abdullah Mohamed Abdelmoemen
Volume
12
Issue
1
Pages
15
publication.type
International
Paper Link
Open Link
Supplementary Materials
Not Available
Abstract
Finding, locating, and resolving software defects takes a lot of time and effort. This paper proposes a hybrid
machine learning model to automate the software testing process. The proposed model combines particle swarm
optimization (PSO) to optimize artificial neural network (ANN) to overcome the local minima and overfitting problems. The
proposed model is compared with different classification algorithms such as: Logistic Regression, K nearest neighbours
(KNN), Decision Tree, Random Forest, Gradient Boosting, AdaBoost, Linear Discriminant Analysis, Quadratic
Discriminant Analysis, Gaussian NB, Support Vector Machine and deep learning neural networks. The effectiveness of the
proposed model is evaluated using four different datasets (CM1, KC1, KC2, and PC1). Datasets have been divided into
training part (70%) and testing part (30%). The proposed model achieved higher accuracy than compared algorithms, while
also reducing time and space complexities.
machine learning model to automate the software testing process. The proposed model combines particle swarm
optimization (PSO) to optimize artificial neural network (ANN) to overcome the local minima and overfitting problems. The
proposed model is compared with different classification algorithms such as: Logistic Regression, K nearest neighbours
(KNN), Decision Tree, Random Forest, Gradient Boosting, AdaBoost, Linear Discriminant Analysis, Quadratic
Discriminant Analysis, Gaussian NB, Support Vector Machine and deep learning neural networks. The effectiveness of the
proposed model is evaluated using four different datasets (CM1, KC1, KC2, and PC1). Datasets have been divided into
training part (70%) and testing part (30%). The proposed model achieved higher accuracy than compared algorithms, while
also reducing time and space complexities.