摘要
为了探究国内深度学习领域的研究进展和热点前沿,综合利用CiteSpace和VOSviewer两款软件对中国知网数据库中2005-2019年深度学习领域相关文献进行科学知识图谱分析,绘制关键词时区图谱、突变检测图谱以及关键词共现权重图谱。通过对所绘知识图谱的分析,可知国内自2014年开始对深度学习领域进行集中研究,理论模型研究热点为卷积神经网络和循环神经网络,应用领域研究热点为计算机视觉、自然语言处理等。其中,深度置信网络应用广泛,至今仍是研究重点。
In order to explore the research progress and hot spot frontier in the field of deep learning in China,this paper comprehen⁃sively used CiteSpace and VOSviewer software to analyze China in the CNKIdatabase.The relevant literature in the field of deep learn⁃ing from 2005 to 2019 were analyzed by scientific knowledge maps,and the keywords time zone map,the mutation word detection map and the co-occurrence weight map were drawn.Based on the analysis of the knowledge maps drawn,it is concluded that since 2014,intensive research has been carried out in the field of deep learning in China.The focus of theoretical model research is convolutional neural network and recurrent neural network.The focus of application is primarily in the fields of computer vision,natural language processing,etc.The application of deep belief network is extensive,and it remains as the focus of research.
作者
徐建国
刘泳慧
刘梦凡
XU Jian-guo;LIU Yong-hui;LIU Meng-fan(College of Computer Science and Engineering,Shandong University of Science and Technology,Qingdao 266590,China)
出处
《软件导刊》
2021年第1期234-237,共4页
Software Guide
基金
山东科技大学研究生科技创新项目(SDKDYC180339)。
