摘要
为了解决在海量web资源中提取出有用军事情报的问题,本文在分析军事情报和互联网信息特点的基础上,设计并实现了一个包括采集、处理、存储与检索的web军事情报挖掘模型,然后提出了一种面向军事情报应用的文本聚类方法,最后通过实验对聚类效果进行了评估,实验结果表明该方法在聚类纯度、准确率、召回率、F-score指标上有不同程度的提升。
In order to extract the useful military intelligence from massive web resources and based on the analysis of military intelligence,characteristics of Internet information,this paper designs and realizes a web mining model of military intelligence including acquisition,processing,storage and retrieval.Then we put forward a kind of text clustering method which oriented military intelligence application and evaluated the clustering effect by experiments ultimately.The experiment results demonstrated that this method have improved the purity of clustering,accuracy,recall rate and F-score index to different extent.
出处
《中国电子科学研究院学报》
北大核心
2015年第5期541-545,共5页
Journal of China Academy of Electronics and Information Technology
