计算机科学 ›› 2018, Vol. 45 ›› Issue (1): 67-72.doi: 10.11896/j.issn.1002-137X.2018.01.010
• CRSSC-CWI-CGrC-3WD 2017 • 上一篇 下一篇
马晓军,郭剑毅,线岩团,毛存礼,严馨,余正涛
MA Xiao-jun, GUO Jian-yi, XIAN Yan-tuan, MAO Cun-li, YAN Xin and YU Zheng-tao
摘要: 实体上下位关系是构建领域知识图谱不可或缺的一种重要的语义关系,传统抽取上下位关系的方法大多不考虑关系的组织。提出一种结合词向量和Bootstrapping的方法来实现领域实体上下位关系的获取与组织。首先,选取旅游领域的种子语料集;然后,采用基于词向量的相似度计算方法对种子集中包含的上下位关系模式进行聚类,筛选出置信度高的模式并对未标注语料进行上下位关系识别,得到候选关系实例,同时选择置信度高的关系实例加入到种子集中,进行下一轮的迭代,直到得到所有的关系实例;最后,根据领域实体上下位关系对的向量偏移并结合领域实体层级关系的特点,采用映射的学习方法进行领域实体层级关系组织。实验结果表明,与传统的方法相比,所提方法的F值提高了近10%。
[1] MILLER G A.WordNet:a lexical database for English[J].Communications of the ACM,1995,38(11):39-41. [2] SHEN D R.SKM:A Schema Matching Model Based on Schema Structure and Known Matching Knowledge[J].Journal of Software,2009,20(2):327-338. [3] HEARST,MARTI A.Automatic acquisition of hyponyms from large text corpora[C]∥Conference on Computational Linguistics.1992:539-545. [4] MANN G S.Fine-grained proper noun ontologies for questionanswering[C]∥The Workshop on Building & Using Semantic Networks.Association for Computational Linguistics,2003. [5] FLEISCHMAN M,HOVY E,ECHIHABI A,et al.Offlinestrategies for online question answering:answering questions before they are asked[C]∥Meeting on Association for Computational Linguistics.Association for Computational Linguistics,2003:1-7. [6] ANDO M,SEKINGE S,ISHIZAKI S.Automatic Extraction of Hyponyms from Newpaper Using Lexicosyntactic Pattern[J].Ipsj Sig Notes,2004,2003:77-82. [7] LIU L,CAO C G,WGNG H T,et al.A Method of Hyponym Acquisition Based on“ isa” Pattern[J].Computer Science,2006,33(9):146-151.(in Chinese) 刘磊,曹存根,王海涛,等.一种基于“是一个”模式的下位概念获取方法[J].计算机科学,2006,33(9):146-151. [8] NAKAYA N,KUREMATSU M,YAMAGUCHI T.A Domain Ontology Development Environment Using a MRD and Text Corpus[J].Casopís Lékar Ceskych,2002,128(37):1166-1169. [9] SUMIDA A,TORISAWA K.Hacking Wikipedia for hyponymy relations acquisition[C]∥International Joint Conference on Natural Language Processing.2008. [10] SUCHANEK FM,KASNECI G,WEIKUM G.Yago:A core of semantic knowledge unifying wordnet and wikipedia[C]∥Proceedings of the Third International Joint Conference on Natural Language Processing.2008:883-888. [11] FAN Q H,ZAN H Y,CHAI Y M,et al(1)hyponym discovery of multiple resource fusion[J].Computer Engineering and Design,2013,34(12):4310-4315.(in Chinese) 范庆虎,昝红英,柴玉梅,等.多资源融合的下位词发现[J].计算机工程与设计,2013,34(12):4310-4315. [12] CARABALLO S A.Automatic Acquisition of a Hypernym-Labeled Noun Hierarchy from Text[C]∥Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics.1999:120-126. [13] BOELLA G,CARO L D.Extracting Definitions and Hypernym Relations Relying on Syntactic Dependencies and Support Vector Machines[C]∥Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics.2013:532-537. [14] ETZIONI O,BANKO M,SODERLAND S,et al.Open information extraction from the web[C]∥International Joint Confe-rence on Artifical Intelligence.Morgan Kaufmann Publishers Inc.,2007:68-74. [15] FANG N M,NON-MEMBER C Y,MEMBER F R.Hyponym extraction from the web by bootstrapping[J].IEEJ Transactions on Electrical & Electronic Engineering,2012,7(7):62-68. [16] KOZAREVA Z,HOVY E.A semi-supervised method to learnand construct taxonomies using the web[C]∥Conference on Empirical Methods in Natural Language Processing.Association for Computational Linguistics.2010:1110-1118. [17] MIKOLOV T,CHEN K,CORRADO G,et al(1)Efficient Estimation of Word Representations in Vector Space[J].Computer Science,2013,3(1):1301-1306. [18] GOLDBERG Y,LEVY O.word2vec Explained:deriving Miko-lov et al(1)’s negative-sampling word-embedding method [J/OL].https://arxiv.org/abs/1402.3722. [19] BENNETT J,GROUT R,PEBAY P,et al(1)Numerically stable, single-pass, parallel statistics algorithms[C]∥IEEE International Conference on Cluster Computing and Workshops.2009:1-8. [20] FU R J,QIN B,LIU T.Exploiting multiple sources for open-domain hypernym discover[C]∥EMNLP.2013:1224-1234. [21] WANG P,HU J,ZENG H J,et al.Improving Text Classification by Using Encyclopedia Knowledge[C]∥IEEE International Conference on Data Mining.IEEE,2007:332-341. |
No related articles found! |
|