5. ESAC. Deep Learning Techniques applied to SW quality assessment - ESAC Trainees
5. Deep Learning Techniques applied to SW quality assessment.
ESA supervisor: Helena Vicente de Castro
Collaborator(s): Fernando Aldea Montero
The ESAC PA provides transversal support to all mission in the Science department. The large amount of data collected on the code quality from the different missions has reached a level where automation and other techniques are required to assess the overall quality of the software products.
Your work will consist in the analysis of the quality metrics to identify areas where Deep Learning can improve the quality assessment of source code, classification of errors, determination of common issues, testing trends, etc. Based on this analysis you will implement Deep Learning (i.e., deep neural networks) to perform one or more of the identified tasks as a demonstrator of the idea. The aim of your project is to derive conclusions and define the way forward on the application of Deep Learning and other Machine Learning techniques to improve source code quality.
Additionally, you will work with the PA team to identify the key aspects the Product Assurance activities applicable to Machine Learning development for scientific data analysis. Based on this, the development cycle should create a new PA approach suitable to this new paradigm comprising new mechanisms for verification and validation.
Project duration: 6 months.
Desirable expertise or programming language:
- Knowledge of Software Engineering process.
- Knowledge of Machine and Deep Learning algorithms.
- Programming languages: Python, Java.
- Knowledge of testing processes and tools.
- Knowledge of CI/CD.
- Knowledge of Matlab/Simulink.
- Knowledge of SW quality assurance practices and techniques.
To apply for this project please fill in an online application form through the following link.