Exploring Software Reusability Metrics with Q&A Forum Data
This work addresses the problem of evaluating software reusability metrics for developers and researchers by leveraging Q&A forum data, though it is incremental as it applies existing NLP techniques to a new domain.
The authors tackled the challenge of analyzing unstructured Q&A forum data to understand software reuse by introducing LANLAN, a method using word embeddings and machine learning to classify messages as problem reports or support requests, achieving an AUROC over 0.9 on data from Eclipse and Bioconductor projects.
Question and answer (Q&A) forums contain valuable information regarding software reuse, but they can be challenging to analyse due to their unstructured free text. Here we introduce a new approach (LANLAN), using word embeddings and machine learning, to harness information available in StackOverflow. Specifically, we consider two different kinds of user communication describing difficulties encountered in software reuse: 'problem reports' point to potential defects, while 'support requests' ask for clarification on software usage. Word embeddings were trained on 1.6 billion tokens from StackOverflow and applied to identify which Q&A forum messages (from two large open source projects: Eclipse and Bioconductor) correspond to problem reports or support requests. LANLAN achieved an area under the receiver operator curve (AUROC) of over 0.9; it can be used to explore the relationship between software reusability metrics and difficulties encountered by users, as well as predict the number of difficulties users will face in the future. Q&A forum data can help improve understanding of software reuse, and may be harnessed as an additional resource to evaluate software reusability metrics.