查询结果:   尹春林,王炜,李彤,何云.基于RNN进行面向主题的特征定位方法[J].计算机应用与软件,2017,34(6):12 - 17,51.
中文标题
基于RNN进行面向主题的特征定位方法
发表栏目
软件技术与研究
摘要点击数
539
英文标题
TOPIC ORIENTED FEATURE LOCALIZATION METHOD BASED ON RNN
作 者
尹春林 王炜 李彤 何云 Yin Chunlin Wang Wei Li Tong He Yun
作者单位
云南大学软件学院 云南 昆明 650500 云南省软件工程重点实验室 云南 昆明 650500    
英文单位
School of Software, Yunnan University,Kunming 650500, Yunnan, China Key Laboratory for Software Engineering of Yunnan Province, Kunming 650500, Yunnan, China    
关键词
软件特征定位 软件演化 深度学习 循环神经网络 面向主题
Keywords
Software feature localization Software evolution Deep learning Recurrent neural network Topic oriented
基金项目
作者资料
尹春林,硕士生,主研领域:软件过程,软件演化,特征定位等。王炜,副教授。李彤,教授。何云,博士生。 。
文章摘要
软件特征定位是软件演化活动得以顺利展开的前提条件。当前特征定位研究的性能仍有待于进一步提高。为了获得较好的性能,在文件夹粒度上获取主题知识,将系统中同一个文件夹下的所有类(class)划分为同一个主题知识类,提出利用深度学习算法——循环神经网络RNN(Recurrent Neural Networks) 进行面向主题的特征定位。同时,在该方法的基础上提出了一种改进的模型。为了使实验结果更具现实意义,与基线方法和其他一些方法相比,将实验数据从10组提高到531组和将检索率从15%缩小到10%,即使在这种情况下,所获得的实验结果,无论是从正面与基线方法相比还是从侧面与目前的一些特征定位方法相比,该方法都获得了不错的性能。
Abstract
Software feature localization is a prerequisite for the smooth development of software evolution. The performance of the current feature location study still needs to be further improved. In order to obtain better performance, get the subject knowledge in the folder granularity was gotten. All the classes under a folder in the system were divided into the same subject knowledge class, This paper proposed a topic-oriented feature locating using depth learning algorithm-Recurrent Neural Networks(RNN). At the same time, an improved model was proposed based on this method. In order to make the experimental results more realistic, compared with the baseline method and other methods, this article will test data from 10 to 531 group and the retrieval rate from 15% to 10%. The experimental results show that this method has better performance than either the baseline method or the feature orientation method.
下载PDF全文