中文专利属性值对抽取技术及应用

J4 ›› 2016, Vol. 38 ›› Issue (04): 800-806.

中文专利属性值对抽取技术及应用

孙东普，朱鸣华，林鸿飞

(大连理工大学计算机科学与技术学院,辽宁大连 116024)

收稿日期:2015-01-27 修回日期:2015-05-15 出版日期:2016-04-25 发布日期:2016-04-25
基金资助:
国家自然科学基金（61202254,61402075）;辽宁省自然科学基金(201202031,201402003)

Chinese patent attributevalue
extraction technology and its application

SUN Dongpu,ZHU Minghua,LIN Hongfei

(School of Computer Science and Technology,Dalian University of Technology,Dalian 116024,China)

Received:2015-01-27 Revised:2015-05-15 Online:2016-04-25 Published:2016-04-25

摘要/Abstract

摘要：

专利信息抽取是专利分析的基础，属性及属性值的识别与抽取是专利信息抽取所要解决的关键问题。目前，在中文专利信息抽取领域针对属性和属性值同步抽取的研究较少。本文以中文专利摘要作为实验语料，运用统计学习知识，提出一种基于条件随机场的抽取方法。该方法将属性和属性值视为命名实体，利用语料训练得到条件随机场模型，从而实现对属性和属性值的抽取；再利用挖掘的关联规则完成属性与属性值匹配。实验结果的准确率、召回率和F值分别是80.8%、81.2%和81.0%，其表明该方法能够高效同步抽取属性和属性值。同时，在抽取结果的基础上，本文完成了对专利的分析和同类专利的比较，体现了本方法的实用价值。

关键词: 属性抽取, 属性值抽取, 中文专利, 条件随机场

Abstract:

Patent information extraction is the foundation of patent analysis, and its attributes and attribute value extraction are important to patent information extraction. However, few studies focus on synchronously extracting attributes and their values in Chinese patent information extraction. Using abstracts of the Chinese patents as corpus, we propose a conditional random fields (CRFs) method based on statistic learning knowledge. Firstly,regarding the attributes and attribute values as named entities,we obtain a CRFs model by training sets, and then use this model to extract attributes and attribute values from the corpus.Secondly, we employ association rules to match the attributes and their values. Experimental results show that the accuracy, recall and Fscore can reach 80.8%, 81.2% and 81.0% respectively.The comparison of the extraction results proves the practical value of the proposal.

Key words: attribute extraction;attribute value extraction;Chinese patent;conditional random fields (CRFs)

孙东普，朱鸣华，林鸿飞. 中文专利属性值对抽取技术及应用[J]. J4, 2016, 38(04): 800-806.

SUN Dongpu,ZHU Minghua,LIN Hongfei. Chinese patent attributevalue
extraction technology and its application [J]. J4, 2016, 38(04): 800-806.

参考文献［15］

［1］	Ding Junjun,Zheng Yanning,Hua Bolin.Survey on attribute extraction at home and abroad［J］.Information Science,2011,29(5):793796.(in Chinese)
［2］	Chen Xu, Peng Zhiyong, Zeng Cheng. A cotraining based method for Chinese patent semantic annotation［C］∥Proc of the 21st ACM International Conference on Information and Knowledge Management,2012:23792382.
［3］	Feng Guangpu,Chen Xu,Peng Zhiyong.A rules and statistical learning based method for Chinese patent information extraction［C］∥Proc of Web Information Systems and Applications Conference (WISA),2011:114118.
［4］	Jin Bo,Teng Hongfei,Shi Yanjun,et al.Chinese patent mining based on sememe statistics and keyphrase extraction［M］∥Advanced Data Mining and Applications.Berlin:Springer Berlin Heidelberg,2007:516523.
［5］	Ding Junjun, Zheng Yanning, Hua Bolin. The academic concept attribute extraction based on rules［J］.Information Studies:Theory & Application,2011,34(12):1014.(in Chinese)
［6］	Tang Wei,Hong Yu, Feng Yanghui, et al. Automatic extraction of the product “attributevalue” pair from the Webpages［J］.Journal of Chinese Information,2013,27(1):2129.(in Chinese)
［7］	Tang W, Hong Y, Feng Y H, et al.Simultaneous product attribute name and value extraction with adaptively learnt templates［C］∥Proc of 2012 International Conference on Computer Science & Service System (CSSS),2012:20212025.
［8］	Li Chunliang,Zhu Yanhui,Xu Yeqiang.Research of attribute word extraction method in Chinese product comment［J］.Computer Engineering,2011,37(12):2628.(in Chinese)
［9］	Ye Zheng, Lin Hongfei, Su Sui，et al. Person attribute extracting based on SVM［J］.Journal of Computer Research and Development,2007,44(Z2):271275.(in Chinese)
［10］	Ghani R,Probst K,Liu Yan,et al.Text mining for product attribute extraction［J］.SIGKDD Explorations,2006,8(1):4148.
［11］	Bing Lidong, Lam Wai, Wong TakLam.Wikipedia entity expansion and attribute extraction from the web using semisupervised learning［C］∥Proc of the 6th ACM International Conference on Web Search and Data Mining,2013:567576.
［12］	Zhang Chunju, Zhang Xueying, Chen Yutian, et al. Extraction of geographical attributevalues in natural language text［C］∥Proc of 2012 International Conference on Environment Science,2012:5159.
［13］	Guo Jianyi,Li Zhen,Yu Zhengtao,et al.Extraction and relation prediction of domain ontology concept instance,attribute and attribute value［J］.Journal of Nanjing University(Natural Sciences),2012,48(4):383389.(in Chinese)
［14］	Lafferty J D, McCallum A, Pereira F C N. Conditional random fields:Probabilistic models for segmenting and labeling sequence data［C］∥Proc of the 18th International Conference on Machine Learning(ICML),2001:282289.
［15］	Li Hang.Statistical learning method［M］.Beijing:Tsinghua University Press,2012.(in Chinese)
	附中文参考文献：
［1］	丁君军,郑彦宁,化柏林.国内外属性抽取研究综述［J］.情报科学,2011,29(5):793796.
［5］	丁君军,郑彦宁,化柏林.基于规则的学术概念属性抽取［J］.情报理论与实践,2011,34(12):1014.
［6］	唐伟,洪宇,冯艳卉,等.网页中商品“属性—值”关系的自动抽取方法研究［J］.中文信息学报,2013,27(1):2129.
［8］	栗春亮,朱艳辉,徐叶强.中文产品评论中属性词抽取方法研究［J］.计算机工程,2011,37(12):2628.
［9］	叶正,林鸿飞,苏绥,等.基于支持向量机的人物属性抽取［J］.计算机研究与发展,2007,44(Z2):271275.
［13］	郭剑毅,李真,余正涛,等.领域本体概念实例,属性和属性值的抽取及关系预测［J］.南京大学学报(自然科学版),2012,48(4):383389.
［15］	李航.统计学习方法［M］.北京:清华大学出版社,2012.

[1]	杜洁, 骆力明, 孙众. 基于ALBERT预训练模型的事件抽取技术研究[J]. 计算机工程与科学, 2023, 45(04): 711-717.
[2]	仇增辉, 赫明杰, 林正奎. 基于深度学习的网购评论命名实体识别方法[J]. 计算机工程与科学, 2020, 42(12): 2287-2294.
[3]	崔丽群，赵越，吴晓冬，魏可飞，刘晨. 一种背景抑制改进的显著性目标检测方法[J]. 计算机工程与科学, 2018, 40(08): 1435-1443.
[4]	王恒升1,2，李熙印2. 机器人自然语言导航的层叠式条件随机场模型[J]. 计算机工程与科学, 2017, 39(08): 1538-1545.
[5]	王文珂1，文雅玫2，蔡喆2. 基于条件随机场模型的数据异常检测算法[J]. J4, 2015, 37(09): 1756-1760.
[6]	吕品1,2,3，钟珞1，蔡敦波2,3，吴云韬2,3. 基于CRF的中文评论有效性挖掘产品特征[J]. J4, 2014, 36(02): 359-366.
[7]	何中市,刘莉,邢欣来,毛小丽. 基于语义角色的中文事件识别[J]. J4, 2013, 35(4): 181-185.
[8]	张朝胜1，郭剑毅1,2，线岩团1,2，余正涛1,2，雷春雅1, 王海雄1. 基于条件随机场的英文产品命名实体识别[J]. J4, 2010, 32(6): 115-117.