Measuring and assesment of well known bad practices in Android applications

Thesis Type: Postgraduate

Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Graduate School of Informatics, Information Systems, Turkey

Approval Date: 2014


Consultant: AYSU BETİN CAN


One of the best ways to make a mobile application usable, reputed and high-scored is attention to the requirements like responsiveness, low memory consumption and stability. To meet these requirements developers must improve their codes by avoiding some bad-practices, which cause "Memory-Leaks", "ANR (Application not responding)" and "Out-of-Memory" to satisfy the user's need and make the Android application robust and usable. In this thesis, I developed a tool that detects a set of bad-practices in Android applications automatically. The tool is applied to source code of 100 open source Android applications. The findings of the tool are used to analyze whether there is a relationship between the user ratings (i.e. the reputation) of the applications with the number and type of bad-practices. To represent reputation, the statistical data of the 100 Android applications that shows their success such as rating and install count is collected from the applications’ official web sites. Another contribution is that, with the aid of the tool developed in this study, developers will be able to find their mistakes in their codes easily or know what may go out wrong when they release their Android applications.