Recovering Android Bad Smells from Android Applications

Abstract

The demand for Android mobile software applications is continuously increasing with the evolution of technology and new enriching features to make the life of people easy and comfortable. The mobile-based software applications are frequently updated as compared to other web and desktop applications. Due to these frequent updating cycles, the developers sometimes make changes in a rush which leads to poor design choices known as antipatterns or code bad smells. Code bad smells degrade the performance of applications and make evolution difficult. The recovery of bad smells from mobile software applications is still at infancy but it is a very important research realm that requires the attention of researchers and practitioners. The results of recovery may be used for comprehension, maintenance, reengineering, evolution and refactoring of these applications. Most state-of-the-art approaches focused on the detection of code bad smells from object-oriented applications and they target only a few code smells. We present a novel approach supplemented with tool support to recover 25 Android code bad smells from Android-specific software applications. We evaluate our approach by performing experiments on 4 open source and 3 industrial Android-specific software applications and measure accuracy using standard metrics.