Building Android applications reliably remains a persistent challenge due to complex dependencies, diverse configurations, and the rapid evolution of the Android ecosystem. This study conducts an empirical analysis of 200 open-source Android projects written in Java and Kotlin to diagnose and resolve build failures. Through a five-phase process encompassing data collection, build execution, failure classification, repair strategy design, and LLM-assisted evaluation, we identified four primary types of build errors: environment issues, dependency and Gradle task errors, configuration problems, and syntax/API incompatibilities. Among the 135 projects that initially failed to build, our diagnostic and repair strategy enabled developers to resolve 102 cases (75.56%), significantly reducing troubleshooting effort. We further examined the potential of Large Language Models, such as GPT-5, to assist in error diagnosis, achieving a 53.3% success rate in suggesting viable fixes. An analysis of project attributes revealed that build success is influenced by programming language, project age, and app size. These findings provide practical insights into improving Android build reliability and advancing AI-assisted software maintenance.

Performance optimization in software engineering is crucial for enhancing user satisfaction and maintaining a competitive advantage. Traditional methods for detecting performance issues – dynamic profiling and static analysis – often fall short in addressing complex dependencies within software architectures. This paper introduces the HyperGraph Neural Network (HGNN), a novel approach that leverages both static and dynamic program analysis to identify and prioritize performance bottlenecks effectively. By analyzing interconnected method call within fundamental patterns, HGNN and its enhanced version, HGNN+, utilize hypergraph neural network techniques to capture and learn dependency relationship features, significantly improving detection accuracy. Our initial testing on the three projects shows that HGNN+, especially when combined with several specific pre-trained models, presents satisfactory results compared to traditional methods.These findings underline HGNN+’s ability to manage complex dependencies and offer a scalable solution for software performance engineering. The benefits observed across multiple initial tests promise a broad applicability for the approach, setting a solid foundation for future research and expansion to more diverse datasets and neural network models, enhancing the reliability and effectiveness of performance optimization detection.