通过标签嵌入从社交标签中挖掘上下位关系  

Mining Hyponymy Relation from Social Tags by Tag Embedding

作　　者：张豹 陈伟荣[1] 张梦易 吴天星 漆桂林[4] ZHANG Bao;CHEN Weirong;ZHANG Mengyi;WU Tianxing;Qi Guilin(The 28th Research Institute of China Electronics Technology Group Corporation,Nanjing 210007,China;School of Computer Science and Engineering,Southeast University,Nanjing 211189,China;School of Computer Science and Engineering,Nanyang Technological University,Singapore 639798,Singapore;Key Laboratory of Computer Network and Information Integration,Southeast University,Nanjing 211189,China)

机构地区：[1]中国电子科技集团公司第二十八研究所,南京210007 [2]东南大学计算机科学与工程学院,南京211189 [3]南洋理工大学计算机科学与工程学院,新加坡639798 [4]东南大学计算机网络和信息集成重点实验室,南京211189

出　　处：《指挥信息系统与技术》2020年第4期64-69,73,共7页Command Information System and Technology

基　　金：装备发展部“十三五”预研课题(31510040201)资助项目。

摘　　要：随着物联网、移动互联网和云计算等技术的飞速发展,数据量急剧增加。获取有价值的信息,尤其是上下位关系知识,成为人工智能领域的热门研究课题。对于分众分类而言,上下位关系识别的目的是识别2个社交标签之间的"is-a"关系。基于此,提出了利用标签嵌入技术从社交标签中识别上下位关系的监督学习方法,采用深度学习算法学习标签嵌入模型,并利用支持向量机算法识别上下位关系。试验结果表明,该方法的准确率和F1值分别达到0.91和0.86,性能优于其他方法。With the rapid development of Internet of Thing,mobile Internet,cloud computing,and other technologies,network data increase dramatically.How to obtain valuable information,especially hyponymy relations,becomes a popular research topic in the field of artificial intelligence.For Folksonomy,the hyponymy relation identification aims to recognize the"is-a"relation between two social tags.A supervised learning method using tag embedding technology to identify the hyponymy relation from social tags is proposed.The method uses a deep learning algorithm to learn tag embedding model,and identifies the hyponymy relation using a support vector machines(SVM)method.Experimental results show that the proposed method outperforms other methods over a labeled dataset with the accuracy and F1-score of 0.91 and 0.86,respectively.

关 键 词：社交标签 上下位关系 词嵌入 分众分类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]