基于深度学习的煤矿领域实体关系抽取研究
|
摘要
关系抽取是构建知识图谱的一个重要过程。为了更好地构建煤矿领域知识图谱,本文对关系抽取的方法进行研究。传统关系抽取方法在训练前多需要人工选取特征、大量标注数据、且需要专业领域的专家辅助、费时费力、且成本较高。本文采用字向量和深度学习相结合的方法对实体间的关系进行抽取,降低数据标注的难度,提高训练效率。实验结果证明使用字向量与深度学习相结合的方法能够较有效地完成煤矿领域实体关系抽取的任务。
Relational extraction is an important process in building knowledge maps. In order to better construct the knowledge map of coal mine field,this paper studies the method of relationship extraction. In this paper,the combination of word vector and deep learning is used to extract the relationship between entities,which reduces the difficulty of data annotation and improves training efficiency. The experimental results prove that the combination of word vector and deep learning can effectively complete the task of extracting entity relations in coal mine field.
Relational extraction is an important process in building knowledge maps. In order to better construct the knowledge map of coal mine field,this paper studies the method of relationship extraction. In this paper,the combination of word vector and deep learning is used to extract the relationship between entities,which reduces the difficulty of data annotation and improves training efficiency. The experimental results prove that the combination of word vector and deep learning can effectively complete the task of extracting entity relations in coal mine field.
引文
[1]GUILLAUMIN M,MENSINK T,VERBEEK J,et al. TagProp:Discriminative metric learning in nearest neighbor models for image auto-annotation[C]//2009 IEEE 12thInternational Conference on Computer Vision. Kyoto,Japan:IEEE,2010:309-316.
[2]邵堃,杨春磊,钱立宾,等.基于模式匹配的结构化信息抽取[J].模式识别与人工智能,2014,27(8):758-768.
[3]CHEN Zhancheng,ZOU Bowei,ZHU Qiaoming,et al. Chinese negation and speculation detection w ith Conditional Random Fields[M]//ZHOU G,LI J,ZHAO D,et al. Natural Language Processing and Chinese Computing. Communications in Computer and Information Science. Berlin/Heidelberg:Springer, 2013,400:30-40.
[4]MANDLE A K,JAIN P,SHRIVASTAVA S K. Protein structure prediction using Support Vector M achine[J]. International Journal on Soft Computing,2012,3(1):67-78.
[5] JIVANI A G. The novel k Nearest Neighbor algorithm[C]//International Conference on Computer Communication and Informatics. Coimbatore,India:IEEE,2013:1-4.
[6] KAMBHATLA N. Combining lexical,syntactic and semantic features w ith maximum entropy models for extracting relations[C]//Proceedings of Association for Computational Linguistics.Barcelona,Spain:ACL,2004:178-181.
[7]陈宇,郑德权,赵铁军.基于Deep Belief Nets的中文名实体关系抽取[J].软件学报,2012,23(10):2572-2585.
[8]AIZERMAN M,BRAVERMAN E,ROZONOER L. Theoretical foundations of the potential function method in pattern recognition learning[J]. Automation&Remote Control,1964,25(6):821-837.
[9]刘克彬,李芳,刘磊,等.基于核函数中文关系自动抽取系统的实现[J].计算机研究与发展,2007,44(8):1406-1411.
[10]ZHOU Peng,SHI Wei,TIAN Jun,et al. Attention-based bidirectional long short-term memory netw orks for relation classification[C]//M eeting of the Association for Computational Linguistics. Berlin,Germany:ACL,2016:207-212.
[11]ZHOU Guobing,WU Jjianxin,ZHANG Chenlin,et al. Minimal gated unit for recurrent neural netw orks[J]. International Journal of Automation&Computing,2016,13(3):226-234.
[2]邵堃,杨春磊,钱立宾,等.基于模式匹配的结构化信息抽取[J].模式识别与人工智能,2014,27(8):758-768.
[3]CHEN Zhancheng,ZOU Bowei,ZHU Qiaoming,et al. Chinese negation and speculation detection w ith Conditional Random Fields[M]//ZHOU G,LI J,ZHAO D,et al. Natural Language Processing and Chinese Computing. Communications in Computer and Information Science. Berlin/Heidelberg:Springer, 2013,400:30-40.
[4]MANDLE A K,JAIN P,SHRIVASTAVA S K. Protein structure prediction using Support Vector M achine[J]. International Journal on Soft Computing,2012,3(1):67-78.
[5] JIVANI A G. The novel k Nearest Neighbor algorithm[C]//International Conference on Computer Communication and Informatics. Coimbatore,India:IEEE,2013:1-4.
[6] KAMBHATLA N. Combining lexical,syntactic and semantic features w ith maximum entropy models for extracting relations[C]//Proceedings of Association for Computational Linguistics.Barcelona,Spain:ACL,2004:178-181.
[7]陈宇,郑德权,赵铁军.基于Deep Belief Nets的中文名实体关系抽取[J].软件学报,2012,23(10):2572-2585.
[8]AIZERMAN M,BRAVERMAN E,ROZONOER L. Theoretical foundations of the potential function method in pattern recognition learning[J]. Automation&Remote Control,1964,25(6):821-837.
[9]刘克彬,李芳,刘磊,等.基于核函数中文关系自动抽取系统的实现[J].计算机研究与发展,2007,44(8):1406-1411.
[10]ZHOU Peng,SHI Wei,TIAN Jun,et al. Attention-based bidirectional long short-term memory netw orks for relation classification[C]//M eeting of the Association for Computational Linguistics. Berlin,Germany:ACL,2016:207-212.
[11]ZHOU Guobing,WU Jjianxin,ZHANG Chenlin,et al. Minimal gated unit for recurrent neural netw orks[J]. International Journal of Automation&Computing,2016,13(3):226-234.