网络文本蕴涵地理信息抽取：研究进展与展望

doi:10.3724/SP.J.1047.2015.00127

摘要/Abstract

摘要：

互联网的普及产生了大量蕴含着丰富地理语义的文本,为地理信息的深度挖掘和知识发现带来了巨大机遇。同时,蕴含地理语义文本的异构性和动态性,使得地理实体的属性数量和种类激增、地理语义关系复杂,对地理信息检索、空间分析和推理、智能化位置服务等提出了严峻的挑战。本文阐述了网络文本蕴含地理信息抽取的技术流程,从地理实体识别、地理实体定位、地理实体属性抽取、地理实体关系构建、地理事件抽取5个方面总结了网络文本蕴含地理信息抽取的进展和关键技术瓶颈,分析了可用于网络文本蕴含地理信息抽取的开放资源,并展望了未来的发展方向。

关键词: 网络文本, 地理信息, 自然语言处理, 信息抽取, 地理定位

Abstract:

Internet generates a plenty of texts which contain abundant geographic semantic information, and bring massive opportunities for deep mining and knowledge discovery. Meanwhile, heterogeneous and dynamic web texts make a surge in the number and type of geographic entity's attributes and the complexity of geographic semantic relations, which present a unprecedented challenge to geographic information retrieval, spatial analysis and reasoning, and intelligent location based services. Firstly, we describe the process of extracting geopgraphic informantion from web texts, summarize the research status and major issues which include geographic entity recognition, locating, attribute extraction, relation construction and event extraction. Secondly, we introduce some popular open sources used for geographic information extraction. Lastly, we discuss and look ahead to the development trends of this domain in future.

Key words: web text, geographic information, natural language processing, infromation extraction, geographical location

余丽, 陆锋, 张恒才. 网络文本蕴涵地理信息抽取：研究进展与展望[J]. 地球信息科学学报, 2015, 17(2): 127-134.DOI:10.3724/SP.J.1047.2015.00127

YU Li,LU Feng,ZHANG Hengcai. Extracting Geographic Information from Web Texts: Status and Development[J]. Journal of Geo-information Science, 2015, 17(2): 127-134.DOI:10.3724/SP.J.1047.2015.00127

图/表 6

图1

表1

表2

表3

表4

表5

参考文献 55

[56]	吴家皋,周凡坤,张雪英.HMM模型和句法分析相结合的事件属性信息抽取[J].南京师大学报(自然科学版),2014,37(1):30-34.
[57]	Jannik S, Michael G, Pavel P.Extraction and exploration of spatio-temporal information[C]. Proceedings of the 6th Workshop on Geographic Information Retrieval, 2010.
[58]	Qiu P Y, Lu F, Zhang H C.Extracting traffic information from Web texts with a D-S evidence theory based approach[C]. 2013 21st International Conference on Geoinformatics, 2013:1-5.
[59]	金澎,吴云芳,俞士汶.词义标注语料库建设综述[J].中文信息学报,2008,22(3):16-23.
[60]	Mihalcea R. The SENSEVAL 3 english lexical sample task[C]. Proceedings of ACL-SIGLEX SENSEVAL 3 worshop, 2004:25-28.
[61]	王跃龙,姬东鸿.汉语树库综述[J].当代语言学,2009,11(1):47-55.
[1]	Sanderson M, Kohler J.Analyzing geographic queries[C]. SIGIR Workshop on Geographic Information Retrieval, 2004.
[2]	Piskorski J, Yangarber R.Information extraction: Past, present and future[C]. Multi-source, Multilingual Information Extraction and Summarization. Berlin Heidelberg: Springer-Verlag, 2013:23-49.
[3]	赵军,刘康,周光有,等.开放式文本信息抽取[J].中文信息学报,2011,25(6):98-110.
[4]	刘振, 张智雄.开放信息抽取技术的现状分析[J].情报杂志,2013,32(11):145-149.
[5]	Oren E, Michael C, Doug D, et al.Unsupervised named-entity extraction from the Web: An experimental study[J]. Artificial Intelligence, 2005,165(1):91-134.
[6]	Joanna B, Erdal K, Fabian M S.Inside YAGO2s: A transparent information extraction architecture[C]. Proceedings of the 22nd International Conference on World Wide Web Companion, 2013:325-328.
[7]	Daniel S W, Raphael H, Fei Wu.Using Wikipedia to bootstrap open information extraction[C]. ACM SIGMOD Record, 2008,37(4):62-68.
[8]	Michele B, Michael J C, Stephen S, et al.Open information extraction from the Web[C]. Proceedings of the 20th International Joint Conference on Artificial Intelligence, 2007:2670-2676.
[9]	Oren E, Anthony F, Janara C, et al.Open information extraction: The second generation[C]. Proceedings of the 22nd International Joint Conference on Artificial Intelligence, 2011:3-10.
[10]	Fei Wu, Daniel S W.Open information extraction using Wikipedia[C]. Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, 2010:118-127.
[11]	Alan A, Alexander L.KrakeN: N-ary facts in open information extraction[C]. Proceedings of the Joint Workshop on Automatic Knowledge Base Construction and Web-scale Knowledge Extraction, 2012:52-56.
[12]	Shi G, Barker K.Extraction of geospatial information on the Web for GIS applications[C]. Proceedings of the 10th IEEE International Conference on Congitive Informatics and Cognitive Computing, 2011:18-20.
[13]	Jones C B, Purves R S.Geographical information retrieval[J]. International Journal of Geographical Information Science, 2008,22(3):219-228.
[14]	Hess B, Gasimov A, Sutanto J.A universal approach that makes legacy online content location-based[C]. Proceedings of the 10th International Conference on Mobile and Ubiquitous Multimedia, 2011:127-133.
[15]	Sundheim B M.Overview of results of the MUC-6 evaluation[C]. Proceedings of the 6th Conference on Message Understanding, 1995:13-31.
[16]	黄德根,岳广玲,杨元生.基于统计的中文地名识别[J].中文信息学报,2002,17(2):36-41.
[17]	Florian A T, Philip D S, Christopher B J.Mining the Web to detect place names[C]. Proceedings of the 2nd International Workshop on Geographic Information Retrieval, 2008:43-44.
[18]	唐旭日,陈小荷,许超,等.基于篇章的中文地名识别研究[J].中文信息学报,2010,24(2):24-32.
[19]	Clare D.Reading geography between the lines: Extracting local place knowledge from text[J]. Spatial Information Theory, 2013,8116:320-337.
[20]	Mónica M, Julián U, Sonia S C, et al.Named entity recognition: Fallacies, challenges and opportunities[J]. Computer Standards & Interfaces, 2013,35(5): 482-489.
[21]	张雪英,闾国年,李伯秋,等.基于规则的中文地址要素解析方法[J].地球信息科学学报,2010,12(1):9-16.
[22]	乐小虬,杨崇俊,刘冬林.空间命名实体的识别[J].计算机工程,2005,31(20):49-53.
[23]	唐旭日,陈小荷,张雪英.中文文本的地名解析方法研究[J].武汉大学学报(信息科学版),2010,35(8): 930-935.
[24]	周俊生,戴新宇,尹存燕,等.基于层叠条件随机场模型的中文机构名自动识别[J].电子学报,2006,34(5):804-809.
[25]	胡万亭,杨燕,尹红风,等.一种基于词频统计的组织机构名识别方法[J].计算机应用研究,2013,30(7):2014-2016.
[26]	李玉森,张雪英,袁正午.面向GIS的地理命名实体识别研究[J].重庆邮电大学学报(自然科学版),2008,20(6):719-724.
[27]	Liu X H, Wei F R, Zhang S D, et al.Named entity recogintion for Tweets[C]. ACM Transactions on Intelligent Systems and Technology, 2013:1-15.
[28]	朱少楠,张雪英,李明,等.基于行政隶属关系树状图的地名消歧方法[J].地理与地理信息科学,2013,29(3):39-42.
[29]	Buscaldi D, Rosso P.A conceptual density-based approach for the disambiguation of toponyms[J]. International Journal of Geographical Information Science, 2008,22(3):301-313.
[30]	Lee L H, Yu Y T, Huang C R.Chinese WordNet domains: Bootstrapping Chinese WordNet with semantic domain labels[C]. Proceedings of PACLIC, 2009:288-296.
[31]	Lieberman M D, Samet H.Multifaceted toponym recognition for streaming news[C]. Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2011:843-852.
[32]	王瑞琴,孔繁胜.无监督词义消歧研究[J].软件学报,2009,20(8):2138-2152.
[33]	刘瑜,袁一泓,张毅.基于认知的模糊地理要素建模——以中关村为例[J].遥感学报,2008,12(2): 370-377.
[34]	Vögele T, Schlieder C, Visser U.Intuitive modelling of place name regions for spatial information retrieval[C]. Spatial Information Theory: Foundations of Geographic Information Science, Berlin Heidelberg: Springer-Verlag, 2003:239-252.
[35]	Steven S, Philip D S, Alia I A, et al.Mining topological relations from the Web[C]. Proceedings of the 19th International Workshop on Database and Expert Systems Application, 2008:652-656.
[36]	Jones C B, Purves R S, Clough P D, et al.Modelling vague places with knowledge from the Web[J]. International Journal of Geographical Information Science, 2008,22(10):1045-1065.
[37]	Durme B V, Qian Ting, Schubert L.Class-driven attribute extraction[C]. Proceedings of the 22nd International Conference on Computational Linguistics, 2008:921-928.
[38]	Putthividhya D P, Hu J L.Bootstrapped named entity recognition for product attribute extraction[C]. Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2011:1557-1567.
[39]	Raju S, Pingali P, Varma V.An unsupervised approach to product attribute extraction[C]. 31st European Conference on IR Research, 2009:6-9.
[40]	Wong T L, Lam W, Wong T S.An unsupervised framework for extracting and normalizing product attributes from multiple Web sites[C]. Proceedings of the 31st Annual International ACM SIGIR Conferenc on Research and Development in Information Retrieval, 2008:35-42.
[41]	贾真,杨宇飞,何大可,等.面向中文网络百科的属性和属性值抽取[J].北京大学学报(自然科学版),2014,50(01):41-47.
[42]	Paşca M, Durme B V, Garera N.The role of documents vs. queries in extracting class attributes from text[C]. Proceedings of the Sixteenth ACM Conference on Information and Knowledge Management, 2007:485-494.
[43]	Ballatore A, Wilson D C, Bertolotto M.Computing the semantic similarity of geographic terms using volunteered lexical definitions[J]. International Journal of Geographical Information Science, 2013,27(10):2099-2118.
[44]	Li W W, Raskin R, Goodchild M F.Semantic similarity measurement based on knowledge mining: An artificial neural net approach[J]. International Journal of Geographical Information Science, 2012,26(8):1415-1435.
[45]	Matsuo Y, Sakaki T, Uchiyama K, et al.Graph-based word clustering using a Web search engine[C]. Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing, 2006:542-550.
[46]	Abreu S C, Bonamigo T L, Vieira R.A review on relation extraction with an eye on Portuguese[J]. Journal of the Brazilian Computer Society, 2013,19(4):553-571.
[47]	张苇如,孙乐,韩先培.基于维基百科和模式聚类的实体关系抽取方法[J].中文信息学报,2012,26(2):75-81.
[48]	Paşca M.Organizing and searching the World Wide Web of facts-step two: Harnessing the wisdom of the crowds[C]. Proceedings of the 16th International Conference on World Wide Web, 2007:101-110.
[49]	张雪英,闾国年.自然语言空间关系及其在GIS中的应用研究[J].地球信息科学,2007,9(6): 77-81.
[50]	乐小虬,杨崇俊,于文洋.基于空间语义角色的自然语言空间概念提取[J].武汉大学学报(信息科学版),2005,30(12):1100-1103.
[51]	朱少楠,张雪英,张春菊.地理空间关系描述的句法模式识别[C].Proceedings of 2010 International Conference on Broadcast Technology and Multimedia Communication,2010:354-357.
[52]	赵妍妍,秦兵,车万翔,等.中文事件抽取技术研究[J].中文信息学报,2008,22(1):3-8.
[53]	许红磊,陈锦秀,周昌乐,等.自动识别事件类别的中文事件抽取技术研究[J].心智与计算,2010,4(1):33-44.
[54]	Shi G, Barker K.Extraction of geospatial information on the Web for GIS applications[C]. 2011 10th IEEE International Conference on Cognitive Informatics & Cognitive Computing, 2011:41-48.
[62]	Ballatore A, Wilson D C, Bertolotto M.A survey of volunteered open geo-knowledge bases in the semantic Web[J]. Quality Issues in the Management of Web Information Intelligent Systems Reference Library, 2013:93-120.
[55]	姜吉发. 一种事件信息抽取模式获取方法[J].计算机工程,2005,31(15):96-98.

名称	下载网址	功能	开发者	开发语言
UIMA	http://uima.apache.org/	NLP工具集	Apache软件基金会	Java、C++
OpenNLP	http://opennlp.apache.org/		Apache软件基金会	Java
LingPipe	http://alias-i.com/lingpipe/index.html		Alias公司	Java
NLTK	http://nltk.org/		团队	Python
GATE	http://gate.ac.uk/		伊利诺斯大学	Java
Mallet	http://mallet.cs.umass.edu/		马萨诸塞大学	Java
StanfordNLP	http://nlp.stanford.edu/		斯坦福大学	Java
GATE	http://gate.ac.uk/	语料标注工具	伊利诺斯大学	Java
Unitex	http://www-igm.univ-mlv.fr/~unitex/index.php?page=0		马恩-拉瓦莱大学	Java、C++
Ellogon	http://www.ellogon.org/		Georgios Petasis	C++、Java、Python
LTP	http://www.ltp-cloud.com/	中文的NLP工具集	哈尔滨工业大学	C#、C++、Java、Ruby、Python
FundanNLP	http://jkx.fudan.edu.cn/nlp/		复旦大学	Java
NLPIR	http://ictclas.nlpir.org/		张华平	C、C++、C#、Java

类型	名称	下载网址	规模	内容
地名数据库	DIVA-GIS Gazetteer	http://www.diva-gis.org/gdata	超过58亿个地名	名称、坐标、地名的层级关系
	OpenStreetMap	http://www.openstreetmap.org	超过20亿个点,2亿个线或面	18大类,地理实体的属性和空间信息
	GeoNames	http://www.geonames.org/	超过1千万个地点	9大类,多语言的地名、坐标、海拔、人口、行政区划、邮编
	Flickr	http://www.flickr.com/	超过140万个地理标签	带有空间位置信息的影像视频数据库
同义词典	TGN	http://www.getty.edu/research/tools/vocabularies/tgn/index.html	超过90万个地点	行政区划和自然地理实体的历史名称、人口、文化、建筑等
	WordNet	http://wordnet.princeton.edu/	超过14万个名词	英文同义词集
	E-HowNet	http://ehownet.iis.sinica.edu.tw/	可在线查询8万个词汇	添加繁体知识中英文同义词集
	同义词词林	http://www.datatang.com/data/42306	超过7万个词汇	中文同义词集,12大类
专题词典	OWTRAD	http://www.ciolek.com/OWTRAD/caravanserais-catalogue-00.html	1051个地名	欧亚非大陆旧贸易路线的客店和驿站的名称、坐标、编码等
专题词典	CH-GIS	http://www.fas.harvard.edu/~chgis/	2513个县名	地点的坐标、层级关系、历史名称、繁体书写、名称来历等

用途	名称	下载网址	规模	语料来源	单位
分词和词性标注	Brown	http://www.hit.uib.no/icame/brown/	超过100万词语	美式英语书面语	布朗大学
	LOB	http://www.hit.uib.no/icame/lobman/lob-cont.html	超过100万词语	英式英语书面语、Brown语料	兰卡斯特大学、奥斯陆大学、挪威计算中心
	BNC	http://www.natcorp.ox.ac.uk/	超过1亿词语	英式英语的书面语和口语	英国牛津出版社等
	现代汉语语料库	http://202.114.40.175:8080/cqs/	在线检索的语料约2千万字	教材、报纸、刊物、图书等	国家语言文字工作委员会
	现代汉语平衡语料库	http://db1x.sinica.edu.tw/cgi-bin/kiwi/mkiwi/kiwi.sh	在线检索限制2000句内	报纸、杂志、BBS、番薯藤网	中国台湾中央研究院
	人民日报切分标注语料库	http://162.105.203.93/icl_groups/corpustagging.asp	已发布语料约200万字	1998年人民日报	北京大学计算语言学研究所
	中文互联网5-gram语料库	http://www.chineseldc.org/doc/CLDC-LAC-2008-001/label.htm	超过8千亿词语	公开访问的网页	Google研究院

用途	名称	下载网址	规模	语料来源	单位
词义消歧	SemCor	http://moin.delph-in.net/SemCor	超过20万词次	WordNet1.6、Brown语料库	普林斯顿大学
	DSO	https://catalog.ldc.upenn.edu/LDC97T12	191个名词和动词,约19万词次	WordNet1.5、Brown语料库、华尔街时报	新加坡国立大学
	SENSEVAL-3^[60]		57个实词,约1.2万词次	WordNet1.7、WordSmyth、BNC	苏塞克斯大学
	STC	http://www.icl.pku.edu.cn/member/wuyf/LanRes.html	约4.6万个动词、2万个名词	现代汉语语义词典、人民日报标注语料库	北京大学

用途	名称	下载网址	规模	语料来源	单位
句法分析	Penn English Treebank	http://www.cis.upenn.edu/~treebank/	约2500个文档	华尔街日报、网页新闻、Brown语料库	宾夕法尼亚大学
	Lancaster-Leeds	http://clwww.esses.ac.uk/w3c/corpus_ling/content/corpora/list/private/LOB/lob.html	约4.5万词语	LOB语料库	兰卡斯特大学
	Penn Chinese Treebank	http://www.cis.upenn.edu/~chinese/ctb.html	超过160万个词语	网页新闻、政府文档、杂志、广播、博客	宾夕法尼亚大学
	汉语依存树库	http://ir.hit.edu.cn/demo/ltp/Sharing_Plan.htm	1万句	Penn Chinese Treebank	哈尔滨工业大学
	中文句结构树库	http://rocling.iis.sinica.edu.tw/CKIP/treebank.htm	超过6万句,可下载1000句	报纸、杂志、BBS、番薯藤网站	中国台湾中央研究院
	Tsinghua Chinese Treebank	http://cslt.riit.tsinghua.edu.cn/~qzhou/eng/Resources.htm	超过100万个词语	文学、学术、新闻、应用的平衡语料文本	清华大学