摘要
随着科学技术的不断发展,各类新型武器装备层出不穷,武器装备信息呈现出的来源更加广泛、形式更加多样、处理更加复杂的态势,如何快速获取、处理、分析武器装备信息对于武器装备的研究论证、开发以及作战运用具有重要意义。论文利用爬虫、文本预处理技术,从门户网站抓取大量武器装备相关信息,构建了武器装备领域的语料库,对基于深度学习的武器装备语料的自动分类方法进行了研究,并提出了基于Attention-CNN的武器装备语料分类方法。实验结果显示该方法准确率为0.9756、召回率为0.9744、F1-score为0.9744,具有较高的分类效果。实验结果表明Attention-CNN方法能够大大提高武器装备相关信息获取、分析效率,对于辅助情报分析人员开展武器装备情报分析工作,为高新武器装备研发、作战运用提供信息保障具有重要意义。
With the continuous development of science and technology,all kinds of new weapons and equipment continue to appear. The information of weapons and equipment presents a situation of more extensive sources,more diverse forms and more complex processing. How to quickly acquire,process and analyze the information of weapon equipment is of great significance for the research,demonstration,development and operational application of weapon equipment. This paper uses crawler and text preprocessing technology to capture a large number of weapon related information from the portal,and constructs a corpus in the field of weapon. Then,the automatic classification method of weapon equipment corpus based on deep learning is studied,and the classification method of weapon equipment corpus based on Attention-CNN is put forward. The experimental results show that the accuracy,recall and F1 score of this method are 0.9756,0.9744 and 0.9744,respectively,which has a high classification effect. The experimental results show that Attention-CNN method can greatly improve the efficiency of weapon equipment related information acquisition and analysis,which is of great significance to assist intelligence analysts to carry out weapon equipment intelligence analysis,and provide information support for high-tech weapon equipment research and development and operational application.
作者
王明乾
邓鹏
倪林
WANG Mingqian;DENG Peng;NI Lin(School of Information and Communication,National University of Defense Technology,Xi'an 710100)
出处
《舰船电子工程》
2022年第2期102-106,111,共6页
Ship Electronic Engineering
关键词
文本分类
语料
武器装备
爬虫
深度学习
text classification
corpus
weapon equipment
crawler
deep learning
