NLPCode: using natural language processing techniques for automated code refactoring

• Grant number: 20/02679-4
• Support type: Scholarships in Brazil - Master
• Effective date (Start): June 01, 2021
• Effective date (End): April 30, 2022
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal researcher: Marcelo Finger
• Grantee: Alan Barzilay
• Home Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 19/07665-4 - Center for Artificial Intelligence, AP.PCPE
• Associated scholarship(s): 21/10104-4 - Function extraction with Transformers, BE.EP.MS

Abstract

Natural Language Processing techniques can be applied to text in general, not only to human language but also for artificial languages such as software code. Code refactoring is a fundamental software engineering technique used both as a quality assurance tool and an important step in code correction and functionality enhancement. In this work, we intend to propose a novel code refactoring model.By utilizing source code as input to our model we wish to obtain automated suggestions of code refactoring in order to achieve better readability and attain good practices in general. The proposed model consists of a neural network that receives a distributed (vectorial) representation of the source code and outputs a representation of the suggested refactored code. This network will be trained based on a repository of refactored code provided through a collaboration with TU Delft Holland. (AU)