Detect and diagnose Code Smell types by Using the Backpropagation Neural Network based on user feedback
Raed Kazem Mohsen, Ahmed Saleem Abbas · Journal of Physics Conference Series · 2020
Abstract Millions of customers rely on smart phone applications for social networking, banking, health, news and many other uses and is also usable anytime, anywhere and in most environmental conditions. However, despite good planning by software engineers and organizations responsible for designing and building applications, the process of building or maintaining the application may be marred by some errors that lead to malfunction of the application or one of its functions, and this is often discovered after a long period, and this affects Opportunities of application success. Therefore, it is necessary to follow up on user feedback on the performance of applications, and to search for tools and methods that contribute to know what users want quickly to save “effort”, “time” and “cost”. This paper discusses some of the main functions of software engineering and the possibility of implementing them in machine learning to detect the presence of design errors “Code Smell “in Android applications with the diagnosis of error type.