一句话：提出互补全局与局部上下文来表示 API 描述，以改进基于 Word2vec 的 API 检索任务。
该论文关注 API 检索中 API 描述与代码片段的语义表示问题。现有基于 Word2vec 的方法通常将单个词或 API 的向量独立聚合为文档表示，未充分考虑词语或 API 之间的上下文依赖，因而难以准确刻画 API 描述和代码片段的语义。标题表明，论文提出在 API 描述表示中互补全局上下文与局部上下文，以更完整地编码文档语义，并服务于 API 检索任务。摘要未给出具体模型结构、数据集与实验指标，因此无法确认其相对基线的量化提升；其核心贡献在于指出独立聚合词向量的局限，并给出结合两类上下文的表示思路。
When being trained on API documentation and tutorials, Word2vec produces vector representations to estimate the relevance between texts and API elements. However, existing Word2vec-based approaches to measure document similarities aggregate Word2vec vectors of individual words or APIs to build the representation of a document as if the words are independent. Thus, the semantics of API descriptions or code fragments are not well represented.
DOI 原文 ·
@inproceedings{paperbot4661,
title = {Complementing global and local contexts in representing API descriptions to improve API retrieval tasks},
author = {Thanh Nguyen and Ngoc Tran and Hung Phan and Trong Nguyen and Linh Truong and Anh Tuan Nguyen and Hoan Anh Nguyen and Tien N. Nguyen},
booktitle = {FSE},
year = {2018},
doi = {10.1145/3236024.3236036}
}