An Automated Approach for Detection and Code Refactoring of Mobile Applications to Enhance Performance

EasyChair Preprint 8778

Abstract

Mobile applications are highly dependent on performance. Performance has become an important aspect on which the quality of applications relies. Detection of problematic code is a way to remove the code smells which automatically improves the performance of the application. If the source code contains bad smells and anti-patterns the performance of the application is compromised. Code smells can directly impact memory, power consumption, and CPU usage. It is identified that the existing literature does not detect 2-3 code smells like “String Concatenation” and Static Views” in android applications. There is also a need to investigate the effect of code smell on performance in mobile applications. Moreover, there is a need to verify empirically that detection has benefits in improving the performance of mobile applications. In this study, we propose an automated approach for the Detection of Code smells in Mobile Applications to enhance Performance. The proposed approach ensures to provide detected code smell with an instance where the smell is detected. Our approach detected the code smell “string concatenation” from the

android applications and 6 other smells. Experiment conducted to show the validity of the approach and the impact of the code smell used. The result of the experiment shows a clear difference in improved processing time without using string concatenation. We evaluated results on open-source applications to detect and refactor the smell and the results show the smell exists in the application. It indicates the instances where the smell was detected and refactored.

Keyphrases: mobile application, performance enhancing, refactoring

@booklet{EasyChair:8778,
  author    = {Hina Shoaib},
  title     = {An Automated Approach for Detection and Code Refactoring of Mobile Applications to Enhance Performance},
  howpublished = {EasyChair Preprint 8778},
  year      = {EasyChair, 2022}}