Herramientas de inteligencia artificial para el proceso de codificación de una aplicación web
Integrating artificial intelligence (AI) into software development presents significant challenges, such as the lack of established standards, the risk of compromising code quality, and resistance to change among development teams. This capstone project addresses these challenges by implementing AI...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| التنسيق: | masterThesis |
| اللغة: | spa |
| منشور في: |
2025
|
| الموضوعات: | |
| الوصول للمادة أونلاين: | https://dspace.unl.edu.ec/jspui/handle/123456789/31842 |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
إضافة وسم | الملخص: | Integrating artificial intelligence (AI) into software development presents significant challenges, such as the lack of established standards, the risk of compromising code quality, and resistance to change among development teams. This capstone project addresses these challenges by implementing AI in developing a web application called DIANA, designed for managing money laundering checklists. Two primary objectives were established: to utilize AI for optimizing the coding process via GitHub Copilot and to assess the learning perception using longitudinal methods and self-evaluations. The Scrum methodology enabled an agile and structured development process, while tools like v0.dev and GitHub Copilot facilitated code generation, review, refactoring, and documentation. Additionally, the AnyLogic simulator was employed to model learning dynamics and analyze the evolution of skills during the project. The results showed that AI not only reduces development times but also improves the learning curve, especially for complex tasks such as unit test creation. Furthermore, longitudinal methods and self-assessments revealed significant progress in learning perception, highlighting key points for improvement and knowledge consolidation. The key contribution of this work is demonstrating how to effectively integrate AI into software development, providing a replicable model to assess its impact on productivity and learning. This approach contributes to the advancement of computer science and software engineering by offering tools and methodologies for the effective adoption of disruptive technologies in real projects. Keywords: Artificial Intelligence (AI), GitHub Copilot, Software Development, Perception of Learning, Longitudinal Methods, AnyLogic Simulator |
|---|