Evaluation of Machine Learning Approaches for Android Energy Bugs Detection With Revision Commits

Abstract

Performances of smartphones are profoundly affected by battery life. Maximizing the amount of usage of energy is essential to extend battery life. However, developers might concentrate more on the functionality of applications while ignoring the energy bugs that drain the battery during the development process. There are no quantitative approaches to detect these energy bugs introduced in this fast-paced development process. In this paper, we employ a system-call-based approach to develop a power consumption model for Android devices. Data that measure the energy consumption of mobile devices under different testing scenarios with the number of triggered system calls are utilized in the model training process. A balanced recursive feature elimination with cross-validation approach is proposed to select and rank the importance of the different system calls. Seven machine learning models are trained over the selected features with cross-validation and hyper-parameter tuning technique, where linear regression with the Lasso regularization outperforms all the other models. Then, the model is evaluated on the data set that measures the energy consumption on different revision history of the selected apps. The results show that the optimized Lasso model could detect energy bugs in the revision history of various applications. Optimization strategies are provided based on the selected features.

Enhancement of GUI Display Error Detection Using Improved Faster R-CNN and Multi-Scale Attention Mechanism

“…One approach is to use deep learning techniques to model the visual information of GUI screenshots and detect display issues [13]. Another approach is to utilize machine learning to automatically detect GUI errors, including widget errors, and analyze the positional relationship between widgets to detect more complex errors [14,15]. Regarding image loss issues, Li et al conducted an empirical study on real-world Android apps and developed a static issue detection tool called TAPIR, which successfully detected previously unknown image-displaying issues [16].…”

Dynamic Testing Techniques of Non-functional Requirements in Mobile Apps: A Systematic Mapping Study

“…The need to test how apps impact battery/power consumption [34], [35], [65], [83], [10], [1], [92], [19], [43], [76], [53] Performance bugs are very diicult to be detected and reproduced [57], [24], [51] Performance testing is a very time-consuming task [43] Lack of tools or methodologies for performance testing [34], [35], [83], [1], [36] The need to detect memory leaks [6] Poor performance apps negatively impact the user experience [6], [51], [90] App performance may vary diferent mobile platforms [81], [76] Developers do not have proper knowledge about performance, therefore they are not careful with this q.c [81], [6], [21], [19] S Developers do not have proper knowledge about security, therefore they are not careful with this q.c [7], [70], [75], [28], [58], [12], [72], [88], [29], [41], [47], [9], [82], [84], [54],…”

“…BB3: Strategy based on ML. Machine learning is used to ind energy ineiciencies [19,36,92]. BB4: Strategy based on execution of predeined sequences of user events.…”

“…MB and FB have been used together in two ways, i.e., EG + FS and EG + FSM. The former one is used to implement a machine learning testing technique [19,36,92] where tests are generated from a mathematical formal speciication of an objective function aimed at guessing energy consumption. The second combination generates test cases to cover speciic paths of an FSM modeling the app behavior.…”

Detecting Wake Lock Leaks in Android Apps Using Machine Learning

“…None of the aforementioned tools and techniques use machine learning to detect wake-lock leaks. However, Zhu et al [34] were the only researchers to use machine learning to detect energy bugs. They used machine learning to train and predict the energy consumption of the changes made in the revision of code commits.…”