arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.16680cs.SEcs.SI
收藏

Stack Overflow 帖子的自动摘要

Automated Summarization of Stack Overflow Posts

Bonan Kou, Muhao Chen, Tianyi Zhang

AI总结:

本文提出基于深度学习的 ASSORT 框架,通过 ASSORT_S 和 ASSORT_IS 两种互补方法解决 Stack Overflow 帖子摘要中标注数据缺乏的问题,在 F1 分数上显著优于现有技术。

摘要:

软件开发人员经常借助 Stack Overflow(SO)来满足其编程需求。鉴于相关帖子数量众多,浏览这些帖子并比较不同解决方案既繁琐又耗时。近期研究提出将 SO 帖子自动摘要为简洁文本,以便于浏览 SO 帖子。然而,这些技术仅依赖信息检索方法或文本摘要启发式规则,不足以应对自然语言的歧义性和复杂性。本文提出了一个基于深度学习的框架,名为 ASSORT,用于 SO 帖子摘要。ASSORT 包含两种互补的学习方法,ASSORT_S 和 ASSORT_IS,以解决 SO 帖子摘要中标注训练数据缺乏的问题。ASSORT_S 旨在直接训练一种新颖的集成学习模型,该模型结合 BERT 嵌入和领域特定特征,以考虑 SO 帖子的独特特性。相比之下,ASSORT_IS 旨在重用预训练模型,同时应对在无训练数据情况下(即零样本学习)的领域偏移挑战。在 F1 分数方面,ASSORT_S 和 ASSORT_IS 分别比六种现有技术至少高出 13% 和 7%。此外,一项人工研究表明,参与者显著偏好 ASSORT_S 和 ASSORT_IS 生成的摘要,而非最佳基线方法,而 ASSORT_S 与 ASSORT_IS 之间的偏好差异较小。

英文摘要:

Software developers often resort to Stack Overflow (SO) to fill their programming needs. Given the abundance of relevant posts, navigating them and comparing different solutions is tedious and time-consuming. Recent work has proposed to automatically summarize SO posts to concise text to facilitate the navigation of SO posts. However, these techniques rely only on information retrieval methods or heuristics for text summarization, which is insufficient to handle the ambiguity and sophistication of natural language. This paper presents a deep learning based framework called ASSORT for SO post summarization. ASSORT includes two complementary learning methods, ASSORT_S and ASSORT_{IS}, to address the lack of labeled training data for SO post summarization. ASSORT_S is designed to directly train a novel ensemble learning model with BERT embeddings and domainspecific features to account for the unique characteristics of SO posts. By contrast, ASSORT_{IS} is designed to reuse pre-trained models while addressing the domain shift challenge when no training data is present (i.e., zero-shot learning). Both ASSORT_S and ASSORT_{IS} outperform six existing techniques by at least 13% and 7% respectively in terms of the F1 score. Furthermore, a human study shows that participants significantly preferred summaries generated by ASSORT_S and ASSORT_{IS} over the best baseline, while the preference difference between ASSORT_S and ASSORT_{IS} was small.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑