论文标题
PTM4TAG：带有预训练模型的堆栈溢出柱的锐化标签建议
PTM4Tag: Sharpening Tag Recommendation of Stack Overflow Posts with Pre-trained Models
论文作者
论文摘要
堆栈溢出通常被视为最具影响力的软件问题答案（SQA）网站，其中包含数百万与编程有关的问题和答案。标签在有效地构造堆栈溢出中的内容方面起着至关重要的作用，对于支持一系列站点操作，例如查询相关内容至关重要。选择不佳的标签通常会引入额外的噪音和冗余，这会导致标签同义词和标签爆炸问题。因此，需要一种可以准确推荐高质量标签的自动标签建议技术，以减轻上述问题。受到自然语言处理（NLP）最新成功的启发，我们提出了PTM4TAG，这是用于堆栈溢出帖子的标签推荐框架，该框架使用具有三重态架构的PTM，该框架模拟了帖子的组件，即标题，描述和独立语言模型的代码。据我们所知，这是SQA站点的标签推荐任务中利用PTM的第一项工作。我们基于五个流行的预培训模型：Bert，Roberta，Albert，Codebert和Bertoverflow评估PTM4TAG的性能。我们的结果表明，PTM4TAG中利用软件工程（SE）特定的PTM Codebert在五个考虑的PTM中取得了最佳性能，并通过平均$ precision@k $，$ beckle@k $和$ f1 $ - $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ - k $ $ $ $ $ $ - k $ $ $ $ $ - k $ $ $ $ $ $ - $ - k k。我们进行了一项消融研究，以量化帖子的组成部分（标题，描述和代码片段）对PTM4TAG性能的贡献。我们的结果表明，标题在预测最相关的标签方面最重要，并且利用所有组件实现了最佳性能。
Stack Overflow is often viewed as the most influential Software Question Answer (SQA) website with millions of programming-related questions and answers. Tags play a critical role in efficiently structuring the contents in Stack Overflow and are vital to support a range of site operations, e.g., querying relevant contents. Poorly selected tags often introduce extra noise and redundancy, which leads to tag synonym and tag explosion problems. Thus, an automated tag recommendation technique that can accurately recommend high-quality tags is desired to alleviate the problems mentioned above. Inspired by the recent success of pre-trained language models (PTMs) in natural language processing (NLP), we present PTM4Tag, a tag recommendation framework for Stack Overflow posts that utilize PTMs with a triplet architecture, which models the components of a post, i.e., Title, Description, and Code with independent language models. To the best of our knowledge, this is the first work that leverages PTMs in the tag recommendation task of SQA sites. We comparatively evaluate the performance of PTM4Tag based on five popular pre-trained models: BERT, RoBERTa, ALBERT, CodeBERT, and BERTOverflow. Our results show that leveraging the software engineering (SE) domain-specific PTM CodeBERT in PTM4Tag achieves the best performance among the five considered PTMs and outperforms the state-of-the-art deep learning (Convolutional Neural Network-based) approach by a large margin in terms of average $Precision@k$, $Recall@k$, and $F1$-$score@k$. We conduct an ablation study to quantify the contribution of a post's constituent components (Title, Description, and Code Snippets) to the performance of PTM4Tag. Our results show that Title is the most important in predicting the most relevant tags, and utilizing all the components achieves the best performance.
