融合表示学习与机器学习的专利科学引文标题自动抽取研究
|
摘要
【目的】针对专利引文类别繁多的问题,研究自动识别其中专利科学引文这一特定类别的方法,进而准确抽取专利科学引文的标题特征项,支撑后续专利科学引文深度分析与挖掘。【方法】引入表示学习方法 Doc2Vec实现专利科学引文整体的语义向量表示,结合机器学习分类方法实现专利科学引文这一特定类别的识别;在此基础上,利用表示学习方法实现专利科学引文标题等内容元数据的语义向量表示,结合机器学习分类方法抽取专利科学引文标题。【结果】在基因领域专利的实验中,专利科学引文的识别精确率达到99.27%,专利科学引文标题抽取精确率达到92.59%,抽取精确率较单纯的机器学习方法提高5.96%。【局限】人工标注训练集较为耗时;对实验数据格式有一定要求。【结论】本文方法在专利科学引文识别和标题抽取上具有良好效果。
[Objective] This paper aims to automatically identify scientific references in patent(SRP), and then extract titles from SRP to support in-depth data mining. [Methods] Firstly, we used the Doc2Vec method to generate vectors for the patent citations. Then, we identified the SRPs with support vector machine(SVM). Third, we created vectors for the metadata(such as titles) of SRP, and extracted titles with SVM. [Results] We examined the proposed method with patent citations from the genetic field. The accuracy of SRP recognition and titles extraction reached 99.27% and 92.59% respectively. The latter was 5.96% higher than those of the traditional methods. [Limitations] Manually tagging the training set was very time consuming, and there are format requirements for the experimental data.[Conclusions] The proposed method could effectively identify and extract patent citations and titles.
[Objective] This paper aims to automatically identify scientific references in patent(SRP), and then extract titles from SRP to support in-depth data mining. [Methods] Firstly, we used the Doc2Vec method to generate vectors for the patent citations. Then, we identified the SRPs with support vector machine(SVM). Third, we created vectors for the metadata(such as titles) of SRP, and extracted titles with SVM. [Results] We examined the proposed method with patent citations from the genetic field. The accuracy of SRP recognition and titles extraction reached 99.27% and 92.59% respectively. The latter was 5.96% higher than those of the traditional methods. [Limitations] Manually tagging the training set was very time consuming, and there are format requirements for the experimental data.[Conclusions] The proposed method could effectively identify and extract patent citations and titles.
引文
[1]Narin F,Hamilton K S,Olivastro D.The Increasing Linkage Between U.S.Technology and Public Science[J].Research Policy,1997,26(3):317-330.
[2]姜霖,王东波.引文元数据的自动发现和标注方法研究--以外文引文为例[J].数据分析与知识发现,2017,1(1):47-54.(Jiang Lin,Wang Dongbo.Automatically Detecting and Tagging Foreign Language Citation Metadata[J].Data Analysis and Knowledge Discovery,2017,1(1):47-54.)
[3]高霞,官建成.非专利引文衍生的科学期刊共被引网络分析[J].科学学研究,2010,28(5):675-680.(Gao Xia,Guan Jiancheng.Co-citation Analysis of Scientific Journal Networks Derived from Non-patent Reference[J].Studies in Science of Science,2010,28(5):675-680.)
[4]Wei W,King I,Lee H M.Bibliographic Attributes Extraction with Layer-upon-Layer Tagging[C]//Proceedings of the 9th International Conference on Document Analysis and Recognition.2007:804-808.
[5]钱建立,吴广茂,蒋路.基于特征相似度的科技论文元数据提取算法研究[J].微电子学与计算机,2008,25(8):129-132.(Qian Jianli,Wu Guangmao,Jiang Lu.Research on Paper Metadata Extraction Algorithm Based on Feature Similarity[J].Microelectronics and Computer,2008,25(8):129-132.)
[6]杨宇,张铭,周宝曜.基于多种规则的课程元数据自动抽取[J].计算机科学,2008,35(3):94-96.(Yang Yu,Zhang Ming,Zhou Baoyao.A Rule-based Metadata Extractor for Learning Materials[J].Computer Science,2008,35(3):94-96.)
[7]Day M Y,Tsai T H,Sung C L,et al.Reference Metadata Extraction Using a Hierarchical Knowledge Representation Framework[J].Decision Support Systems,2007,43(1):152-167.
[8]Cortez E,Silva A S D,Mesquita F,et al.FLUX-CIM:Flexible Unsupervised Extraction of Citation Metadata[C]//Proceedings of the 7th ACM/IEEE-CS Joint Conference on Digital Libraries.ACM,2007:215-224.
[9]Seymore K,McCallum A,Rosenfeld R.Learning Hidden Markov Model Structure for Information Extraction[C]//Proceedings of the 1999 AAAI Workshop on Machine Learning for Information Extraction.1999:37-42.
[10]Nanba H,Anzen N,Okumura M.Automatic Extraction of Citation Information in Japanese Patent Applications[J].International Journal on Digital Libraries,2008,9(2):151-161.
[11]Han H,Giles C L,Manavoglu E,et al.Automatic Document Metadata Extraction Using Support Vector Machines[C]//Proceedings of the 2003 Joint Conference on Digital Libraries.IEEE,2003:37-48.
[12]张铭,银平,邓志鸿,等.SVM+BiHMM:基于统计方法的元数据抽取混合模型[J].软件学报,2008,19(2):358-368.(Zhang Ming,Yin Ping,Deng Zhihong,et al.SVM+BiHMM:A Hybrid Statistic Model for Metadata Extraction[J].Journal of Software,2008,19(2):358-368.)
[13]蒋新.英美学术文献的几种主要引文方式[J].图书与情报,2003(3):26-30.(Jiang Xin.Several Main Quotation Ways in British-American Academic Documents[J].Library and Information,2003(3):26-30.)
[14]Le Q,Mikolov T.Distributed Representations of Sentences and Documents[C]//Proceedings of the 2014 International Conference on Machine Learning.2014:1188-1196.
[15]Mikolov T,Chen K,Corrado G,et al.Efficient Estimation of Word Representations in Vector Space[OL].arXiv Preprint.arXiv:1301.3781.
[16]Hinton G E.Learning Distributed Representations of Concepts[C]//Proceedings of the 8th Annual Conference of the Cognitive Science Society.1986:1-12.
[17]于政.基于深度学习的文本向量化研究与应用[D].上海:华东师范大学,2016.(Yu Zheng.The Study and Application of Text Embeddings with Deep Learning Technique[D].Shanghai:East China Normal University,2016.)
[2]姜霖,王东波.引文元数据的自动发现和标注方法研究--以外文引文为例[J].数据分析与知识发现,2017,1(1):47-54.(Jiang Lin,Wang Dongbo.Automatically Detecting and Tagging Foreign Language Citation Metadata[J].Data Analysis and Knowledge Discovery,2017,1(1):47-54.)
[3]高霞,官建成.非专利引文衍生的科学期刊共被引网络分析[J].科学学研究,2010,28(5):675-680.(Gao Xia,Guan Jiancheng.Co-citation Analysis of Scientific Journal Networks Derived from Non-patent Reference[J].Studies in Science of Science,2010,28(5):675-680.)
[4]Wei W,King I,Lee H M.Bibliographic Attributes Extraction with Layer-upon-Layer Tagging[C]//Proceedings of the 9th International Conference on Document Analysis and Recognition.2007:804-808.
[5]钱建立,吴广茂,蒋路.基于特征相似度的科技论文元数据提取算法研究[J].微电子学与计算机,2008,25(8):129-132.(Qian Jianli,Wu Guangmao,Jiang Lu.Research on Paper Metadata Extraction Algorithm Based on Feature Similarity[J].Microelectronics and Computer,2008,25(8):129-132.)
[6]杨宇,张铭,周宝曜.基于多种规则的课程元数据自动抽取[J].计算机科学,2008,35(3):94-96.(Yang Yu,Zhang Ming,Zhou Baoyao.A Rule-based Metadata Extractor for Learning Materials[J].Computer Science,2008,35(3):94-96.)
[7]Day M Y,Tsai T H,Sung C L,et al.Reference Metadata Extraction Using a Hierarchical Knowledge Representation Framework[J].Decision Support Systems,2007,43(1):152-167.
[8]Cortez E,Silva A S D,Mesquita F,et al.FLUX-CIM:Flexible Unsupervised Extraction of Citation Metadata[C]//Proceedings of the 7th ACM/IEEE-CS Joint Conference on Digital Libraries.ACM,2007:215-224.
[9]Seymore K,McCallum A,Rosenfeld R.Learning Hidden Markov Model Structure for Information Extraction[C]//Proceedings of the 1999 AAAI Workshop on Machine Learning for Information Extraction.1999:37-42.
[10]Nanba H,Anzen N,Okumura M.Automatic Extraction of Citation Information in Japanese Patent Applications[J].International Journal on Digital Libraries,2008,9(2):151-161.
[11]Han H,Giles C L,Manavoglu E,et al.Automatic Document Metadata Extraction Using Support Vector Machines[C]//Proceedings of the 2003 Joint Conference on Digital Libraries.IEEE,2003:37-48.
[12]张铭,银平,邓志鸿,等.SVM+BiHMM:基于统计方法的元数据抽取混合模型[J].软件学报,2008,19(2):358-368.(Zhang Ming,Yin Ping,Deng Zhihong,et al.SVM+BiHMM:A Hybrid Statistic Model for Metadata Extraction[J].Journal of Software,2008,19(2):358-368.)
[13]蒋新.英美学术文献的几种主要引文方式[J].图书与情报,2003(3):26-30.(Jiang Xin.Several Main Quotation Ways in British-American Academic Documents[J].Library and Information,2003(3):26-30.)
[14]Le Q,Mikolov T.Distributed Representations of Sentences and Documents[C]//Proceedings of the 2014 International Conference on Machine Learning.2014:1188-1196.
[15]Mikolov T,Chen K,Corrado G,et al.Efficient Estimation of Word Representations in Vector Space[OL].arXiv Preprint.arXiv:1301.3781.
[16]Hinton G E.Learning Distributed Representations of Concepts[C]//Proceedings of the 8th Annual Conference of the Cognitive Science Society.1986:1-12.
[17]于政.基于深度学习的文本向量化研究与应用[D].上海:华东师范大学,2016.(Yu Zheng.The Study and Application of Text Embeddings with Deep Learning Technique[D].Shanghai:East China Normal University,2016.)