###

计算机系统应用英文版:2019,28(9):209-214

View/Add Comment 过刊浏览高级检索 HTML

←前一篇 | 后一篇→

码上扫一扫！

下载全文

基于注意力机制的Java API序列推荐方法

张睿峰^1,2, 王鹏程^1,2, 吴鸣^1,2, 徐云^1,2

(1.中国科学技术大学计算机科学与技术学院, 合肥 230026;2.安徽省高性能计算重点实验室, 合肥 230026)

Java API Sequence Recommendation Method Based on Attention Mechanism

ZHANG Rui-Feng^1,2, WANG Peng-Cheng^1,2, WU Ming^1,2, XU Yun^1,2

(1.School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China;2.Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China)

摘要

图/表

参考文献

相似文献

本文已被：浏览 1245次下载 1657次
Received:February 26, 2019 Revised:March 14, 2019

中文摘要: 软件开发者在软件代码中如何正确使用API和API序列（APIs），是一个需要学习的困难过程.于是面对不熟悉函数库或像Github那样包含大量APIs的代码仓库，需要一些推荐工具或系统辅助开发者的APIs使用.目前我们所知最好的方法DeepApi能较好理解用户的查询语义，但基于RNN的模型存在问题：（1）没有考虑每个单词的权重；（2）将输入序列压缩为一个固定长度的向量，损失了较多有用信息；（3）句子过长会使关键信息丢失.为此，本文使用了一种基于注意力机制的模型，可以区分每个单词的重要程度并解决长查询输入所产生的长距离依赖问题.我们从Github上面爬取了649个Java开源项目，经过处理得到有114 364对注释-API序列的训练集.实验结果表明我们的方法比DeepApi方法对于BLUE指标在Top1、Top5、Top10上均能提升约20%以上.

中文关键词: API序列推荐注意力机制深度学习

Abstract:It is a difficult process for developers to use API and API sequences (APIs) correctly in software development. When developers are faced with unfamiliar function libraries or code repositories like Github that contains a large number of APIs, they need assistance of some recommendation tools or system. To the best of our knowledge, DeepApi can better understand the semantics of user’s query, but the RNN-based model has some problems: (1) it does not consider the weight of each word, (2) the input sequence is compressed into a fixed length vector, which loses much useful information, (3) long sentences lead to loss of key information. Therefore, this study uses a model based on attention mechanism to distinguish the importance of each word and solve the problem of long-distance dependence caused by long query input. We crawled 649 Java open source projects from Github and processed them to get a training set of 114 364 pairs of annotation-API sequences. The experimental results show that the proposed method can increase BLUE index by more than about 20% compared with DeepApi method on Top1, Top5, and Top10.

keywords: API sequences recommendation attention mechanisms deep learning

文章编号： 中图分类号： 文献标志码：

基金项目:国家自然科学基金面上项目（61672480）

Author Name	Affiliation	E-mail
ZHANG Rui-Feng	School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China
WANG Peng-Cheng	School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China
WU Ming	School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China
XU Yun	School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China	xuyun@ustc.edu.cn

Author Name	Affiliation	E-mail
ZHANG Rui-Feng	School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China
WANG Peng-Cheng	School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China
WU Ming	School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China
XU Yun	School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, China Key Laboratory of High Performance Computing of Anhui Province, Hefei 230026, China	xuyun@ustc.edu.cn

引用文本：
张睿峰,王鹏程,吴鸣,徐云.基于注意力机制的Java API序列推荐方法.计算机系统应用,2019,28(9):209-214
ZHANG Rui-Feng,WANG Peng-Cheng,WU Ming,XU Yun.Java API Sequence Recommendation Method Based on Attention Mechanism.COMPUTER SYSTEMS APPLICATIONS,2019,28(9):209-214