Titulo Estágio
Software Project Analysis by Application of Machine Learning to Project Repositories
Áreas de especialidade
Engenharia de Software
Sistemas Inteligentes
Local do Estágio
DEI
Enquadramento
Recent advances in machine learning, leading to the widespread adoption of deep learning neural networks to mass market applications, has raised a renewed interest in the application of these technologies in very diverse domains.
One of these domains is the software development process itself, as it is a typical scenario where terabytes of semi-structured software project information is kept in project repositories, such as git.
Objetivo
The goal of this Master Thesis is to study the application of machine Learning techniques to software project repositories in order to identify factors that might have a significant in the project success.
Plano de Trabalhos - Semestre 1
First Semester:
1) Getting awareness of the technologies to use.
2) Selection of data visualization tools for understanding the results.
3) Literature survey to identify similar research works.
4) Definition of the experimental setup and research plan.
5) Elaboration of the intermediate report
An actual real-world project repository will be used, with collaboration from an external software company.
Plano de Trabalhos - Semestre 2
Second Semester:
1) Update of the intermediate report with the feedback from the jury.
2) In this semester we shall perform an iterative approach in three-to-four week sprints:
i.a data analysis architecture will be designed, knowledge gathered and a new/improved experiment performed on the data.
ii.Update of the final report with the incremental results obtained.
3.Writing a scientific research paper with the most relevant results.
4.Elaboration of the final version of the Master Thesis (consolidation of the work performed so far).
Condições
n/a
Orientador
Mário Alberto da Costa Zenha Rela
mzrela@dei.uc.pt 📩