A Comprehensive Study on the Use of Word Embedding Models in Software Engineering Domain

TL;DR

This paper provides a comprehensive analysis of how word embedding models are used in software engineering, comparing methods, training strategies, and discussing challenges to improve semantic representations of software artifacts.

Contribution

It systematically reviews 181 studies, offering a unified view of current practices and identifying challenges in applying word embeddings to software engineering tasks.

Findings

Systematic overview of WE applications in SE

Comparison of WE with traditional semantic methods

Identification of challenges in adopting WE in SE

Abstract

Word embedding (WE) techniques are advanced textual semantic representation models oriented from the natural language processing (NLP) area. Inspired by their effectiveness in facilitating various NLP tasks, more and more researchers attempt to adopt these WE models for their software engineering (SE) tasks, of which semantic representation of software artifacts such as bug reports and code snippets is the basis for further model building. However, existing studies are generally isolated from each other without comprehensive comparison and discussion. This not only makes the best practice of such cross-discipline technique adoption buried in scattered papers, but also makes us kind of blind to current progress in the semantic representation of SE artifacts. To this end, we decided to perform a comprehensive study on the use of WE models in the SE domain. 181 primary studies published in mainstream software engineering venues are collected for analysis. Several research questions related to the SE applications, the training strategy of WE models, the comparison with traditional semantic representation methods, etc., are answered. With the answers, we get a systematical view of the current practice of using WE for the SE domain, and figure out the challenges and actions in adopting or developing practical semantic representation approaches for the SE artifacts used in a series of SE tasks.