An Evaluation of the State-of-the-Art Testing Algorithms for Android Mobile Applications

In the past decade, the smartphone industry has blossomed into one of the biggest markets in modern technology. Over 6 million people own smartphones in 2021, so the rapid detection and elimination of any errors are important. Modern applications tend to have tens of thousands of lines of code, meaning that manual testing is both unreliable and impractical. To ensure that our applications are secure and safe to use, we need to create algorithms to assist us in the testing process. The majority of research in the automated testing field has been directed towards the Android platform. It has the largest share of the current mobile market and is open-source in nature, making it an appealing target for research. In recent years, several automated testing techniques have been developed to debug Android applications. These algorithms use a variety of methods to automate testing, ranging from randomization to machine learning. This paper will thoroughly compare different algorithms used to automate Android testing and will evaluate the most efficient methods. The criteria for evaluating each technique prioritize coverage, error detection ability, and computational efficiency