摘要
为了提高软件复用过程中构件检索的效率,分析了软件构件分类技术的优缺点以及构件特征,从构件刻面信息的角度,采用卷积神经网络技术,提出一种基于卷积神经网络的构件分类策略;利用卷积神经网络对构件刻面特征进行提取,减少人为因素,提高刻面信息提取精确性,并训练出基于卷积神经网络的构件分类模型,通过具体的实验,来论证该模型的准确性,以达到提高构件检索效率的目的。
To improve the efficiency of component retrieval in software reuse process, advantages and disadvantages of software component classification technology and component features are analyzed. From the aspect of component facet-ing information, convolution neural network technology is used to propose a component based on convolution neural network. The convolutional neural network is used to extract the facet features, reduce the human factors, improve the accuracy of faceted information extraction, and train the component classification model based on convolution neural network. Through the specific experiments, the model is demonstrated to improve the efficiency of component retrieval.
作者
张富为
杨秋翔
宋超峰
ZHANG Fuwei;YANG Qiuxiang;SONG Chaofeng(College of Software, North University of China, Taiyuan 030051, China)
出处
《计算机工程与应用》
CSCD
北大核心
2019年第8期201-207,共7页
Computer Engineering and Applications
基金
山西省科技公关基金(No.20090322004)
关键词
软件复用
软件构件
卷积神经网络
构件分类
构件检索
software reuse
software component
convolution neural network
component classification
component retrieval
