FACOS: Finding API Relevant Contents on Stack Overflow with Semantic and Syntactic Analysis

Abstract

Collecting API examples, usages, and mentions relevant to a specific API method over discussions on venues such as Stack Overflow is not a trivial problem. It requires efforts to correctly recognize whether the discussion refers to the API method that developers/tools are searching for. The content of the thread, which consists of both text paragraphs describing the involvement of the API method in the discussion and the code snippets containing the API invocation, may refer to the given API method. Leveraging this observation, we develop FACOS, a context-specific algorithm to capture the semantic and syntactic information of the paragraphs and code snippets in a discussion. FACOS combines a syntactic word-based score with a score from a predictive model fine-tuned from CodeBERT. FACOS beats the state-of-the-art approach by 13.9% in terms of F1-score.

On the Effectiveness of Pretrained Models for API Learning

“…Firstly, The CodeBERT developers make use of both bimodal instances of NL-PL pairs (i.e., code snippets and function-level comments or documentations) and a large amount of available unimodal codes [23]. In addition, the developers have pre-trained CodeBERT using a hybrid objective function, which includes masked language modeling [7] and replaced token detection [6].…”