Vol 5, No 1 (2020)

Integration of On-Device Machine Learning (TensorFlow Lite)

Author: Kunal Singh , Shivam Dubey

Abstract:Android applications have become a core component of modern software ecosystems, serving millions of users across diverse domains such as healthcare, finance, education, and entertainment. Ensuring the quality and reliability of Android applications is therefore critical. Mutation testing is recognized as one of the most powerful fault-based software testing techniques, as it evaluates the effectiveness of test suites by introducing small artificial faults, known as mutants, into the program. However, direct application of traditional mutation testing techniques to Android apps faces several challenges including high computational cost, large number of generated mutants, platform-specific features, and event-driven architectures. This paper presents a comprehensive review of efficient mutation testing methods for Android applications. We discuss Android-specific mutation operators, optimization strategies such as selective mutation, higher-order mutation, and mutant reduction, as well as tool support for practical adoption. The paper also compares existing approaches in terms of efficiency, fault detection capability, and scalability. Finally, current research challenges and future directions are highlighted to guide further work in this area.

Keywords: Android Testing, Mutation Testing, Mobile Application Testing, Test Suite Evaluation, Software Quality

Full Issue

View or download the full issue 38-50 PDF

Table of Contents