Bridging Semantic Gaps between Natural Languages and APIs with Word Embedding

Abstract

Developers increasingly rely on text matching tools to analyze the relation between natural language words and APIs. However, semantic gaps, namely textual mismatches between words and APIs, negatively affect these tools. Previous studies have transformed words or APIs into low-dimensional vectors for matching; however, inaccurate results were obtained due to the failure of modeling words and APIs simultaneously. To resolve this problem, two main challenges are to be addressed: the acquisition of massive words and APIs for mining and the alignment of words and APIs for modeling. Therefore, this study proposes Word2API to effectively estimate relatedness of words and APIs. Word2API collects millions of commonly used words and APIs from code repositories to address the acquisition challenge. Then, a shuffling strategy is used to transform related words and APIs into tuples to address the alignment challenge. Using these tuples, Word2API models words and APIs simultaneously. Word2API outperforms baselines by 10%-49.6% of relatedness estimation in terms of precision and NDCG. Word2API is also effective on solving typical software tasks, e.g., query expansion and API documents linking. A simple system with Word2API-expanded queries recommends up to 21.4% more related APIs for developers. Meanwhile, Word2API improves comparison algorithms by 7.9%-17.4% in linking questions in Question&Answer communities to API documents.

Software defect prediction with imbalanced distribution by radius‐synthetic minority over‐sampling technique

“…At present, in order to ensure the quality of software, many scholars explore the application of machine learning in this field. [25][26][27][28][29][30] There are also many studies focusing on software defect prediction, [31][32][33][34] which mainly apply machine learning to two parts; one is Within-Project Defect Prediction (WPDP), and the other is Cross-Project Defect Prediction (CPDP).…”

Data‐driven approach to application programming interface documentation mining: A review

“…To evaluate the effectiveness of data-driven approach in ADM, various evaluation metrics have been widely used in ADM (Gu et al, 2016;Huang et al, 2018;Li, Jiang, et al, 2018;Raghothaman et al, 2016;Rahman et al, 2016;Yuan et al, 2019;Zhang, Niu, Keivanloo, & Zou, 2018).…”

“…For example, when solving the mismatched problems, X. Ye, Shen, et al (2016) and Huang et al (2018) utilize MRR and MAP (Mean Average Precision), Rahman et al (2016) use MRR, MAP, Top@k and R@k (Recall at position k), and Li, Jiang, et al (2018) choose to use MRR, MAP, and FP (First Rank). The aforementioned approaches are evaluated by using inconsistent metrics.…”

BLESER: Bug Localization Based on Enhanced Semantic Retrieval

“…Specifically, we used an abstract syntax tree (AST) based model to retrieve the functional semantics of methods. The semantics of bug reports were retrieved by state-of-the-art word embedding models which have been proved to be effective in representing textual semantics in various natural language tasks [3], [6], [7]. Then, we used a deep learning model that leverages two kinds of semantic features of methods and bug reports, to learn unified features from methods and bug reports to automatically locate buggy methods for a given bug report.…”