Faculty of Computing
Permanent URI for this communityhttps://repository.nileuniversity.edu.ng/handle/123456789/13
Browse
5 results
Search Results
Item Test Case Generation Approach for Android Applications using Reinforcement Learning(Engineering, Technology & Applied Science Research, 2024-04-27) Moussa Mahamat Boukar; Muhammed Aliyu Suleiman; Ibrahim Anka SalihuMobile applications can recognize their computational setting and adjust and respond to actions in the context. This is known as context-aware computing. Testing context-aware applications is difficult due to their dynamic nature, as the context is constantly changing. Most mobile testing tools and approaches focus only on GUI events, adding to the deficient coverage of applications throughout testing. Generating test cases for various context events in Android applications can be achieved using reinforcement learning algorithms. This study proposes an approach for generating Android application test cases based on Expected State-Action-Reward-State-Action (E-SARSA), considering GUI and context events for effective testing. The proposed method was experimentally evaluated on eight Android applications, showing 48- 96% line of code coverage across them, which was higher than Q-testing and SARSA.Item Mutation Testing Techniques for Android Applications: A Comparative Study(International Conference on Multidisciplinary Engineering and Applied Science, 2023-02-02) Ibrahim Anka SalihuMobile applications are becoming increasingly used to achieve various computing needs. Hence, it is essential to guarantee quality of the applications. Software testing has been the main activity for ensuring the quality of software, however, it is a critical and costly activity. Furthermore, code coverage is commonly used as metric for verifying the effectiveness of testing technique. However, numerous researchers and experts claimed that considering code coverage only to ascertain the testing quality is not enough to ensure quality of apps. In view of this, mutation testing has been introduced to complement the procedure by assuring that apps behave as expected and are released free of faults. Several techniques/tools were proposed based on different syntax and mutation operators. This study presents a comparative study of six (6) mutation techniques/tools for Android applications to highlight their strengths and weakness which will give an insight to researchers on the directions for further researchItem Review of Machine Learning Techniques For Class Imbalance Medical Dataset(International Conference on Multidisciplinary Engineering and Applied Science, 2023-02-02) Ibrahim Anka SalihuData imbalance threatens a medical dataset where the dominant class is typically viewed as unfavorable. In contrast, the minority class is supposed to be the positive one, affecting the machine learning prediction performance. This aims to examine how resampling strategies in Machine Learning(ML) have recently been used in medical data sets. Many researchers used the preprocessing stage's data-level approach to resample the imbalanced medical data. Thirty-two sources were reviewed in which data level techniques of balancing the imbalanced data were applied to medical datasets spanning 2018 to 2023, with oversampling methods outperforming the under-sampling methods.Item Comparative Analysis of Fully Automated Testing Techniques for Android Applications(International Conference on Multidisciplinary Engineering and Applied Sciences (ICMEAS-2023), 2023-02-02) Ibrahim Anka SalihuThe software testing community has continued to research and develop new ways of testing Android mobile applications to ensure an application function as expected and serves its purpose. Given that every testing technique/tool has its strengths and weaknesses. This study aims to present a comparative analysis of fully automated techniques which automatically generate and execute test cases concurrently during runtime. We selected 10 fully automated techniques published from 2013 to 2023 techniques to identify the similarities and differences that exist among them. We clearly define the comparative criteria (such as the exploration method, systematic method, termination criterion, extraction criterion, and scheduling method) used for comparing the selected techniques. The analysis shows that most fully automated techniques adopt an active learning exploration approach for exploring the application under test (AUT). We also observed that only the techniques utilizing the active learning approach are capable of modelling and abstracting the graphic user interface (GUI) of the AUT, others randomly select events to be fired into the AUT.Item Reinforcement Learning for Testing Android Applications(International Conference on Multidisciplinary Engineering and Applied Science, 2023-02-02) Moussa Mahamat Boukar; Muhammed Aliyu Suleiman; Ibrahim Anka SalihuThis paper offers a review of current research studies that use reinforcement learning (RL) to test Android applications. The primary purpose of this study is to simplify future research by collecting and investigating the current state of Android app testing approaches using the RL technique. We provide a well-defined criterion comprising of seven key points. The key points are: addressed problems, reasons for using the RL technique, RL algorithms, supported events, testing techniques, validation, and evaluation methods. In the literature, we have analyzed various techniques to evaluate their efficiency. This study showed that model-based testing is the most commonly used testing technique. Q-learning is the best algorithm in terms of predictive accuracy. We identified that code coverage is the most widely used evaluation metric and comparison with other tools and techniques is the preferred validation approach.