A neural joint model for entity and relation extraction from biomedical text

BMC Bioinformatics

Research Article

Meishan Zhang¹  Guohong Fu¹  Fei Li²  Donghong Ji² 

[1] School of Computer Science and Technology, Heilongjiang University, Xuefu Road, Harbin, China;School of Computer, Wuhan University, Bayi Road, Wuhan, China;

关键词: Biomedical text;    Entity recognition;    Relation extraction;    Neural network;    Joint model;   

DOI  :  10.1186/s12859-017-1609-9

 received in 2016-11-01, accepted in 2017-03-23,  发布年份 2017

来源: Springer

【 摘 要 】

BackgroundExtracting biomedical entities and their relations from text has important applications on biomedical research. Previous work primarily utilized feature-based pipeline models to process this task. Many efforts need to be made on feature engineering when feature-based models are employed. Moreover, pipeline models may suffer error propagation and are not able to utilize the interactions between subtasks. Therefore, we propose a neural joint model to extract biomedical entities as well as their relations simultaneously, and it can alleviate the problems above.ResultsOur model was evaluated on two tasks, i.e., the task of extracting adverse drug events between drug and disease entities, and the task of extracting resident relations between bacteria and location entities. Compared with the state-of-the-art systems in these tasks, our model improved the F1 scores of the first task by 5.1% in entity recognition and 8.0% in relation extraction, and that of the second task by 9.2% in relation extraction.ConclusionsThe proposed model achieves competitive performances with less work on feature engineering. We demonstrate that the model based on neural networks is effective for biomedical entity and relation extraction. In addition, parameter sharing is an alternative method for neural models to jointly process this task. Our work can facilitate the research on biomedical text mining.

【 授权许可】

CC BY   
© The Author(s) 2017

【 图 表 】

12864_2015_2304_Article_IEq11.gif

12864_2015_2129_Article_IEq18.gif

12864_2017_3920_Article_IEq2.gif

12864_2017_3920_Article_IEq1.gif

12894_2016_184_Article_IEq3.gif

12864_2015_2129_Article_IEq12.gif

12864_2017_3661_Article_IEq1.gif

12864_2015_2129_Article_IEq10.gif

12864_2015_2129_Article_IEq9.gif

12864_2015_2129_Article_IEq8.gif

12864_2015_2055_Article_IEq29.gif

12864_2017_3990_Article_IEq6.gif

12864_2015_2192_Article_IEq21.gif

12864_2017_3610_Article_IEq1.gif

12864_2015_2055_Article_IEq26.gif

12888_2016_877_Article_IEq15.gif

12864_2017_3687_Article_IEq1.gif

12864_2016_2791_Article_IEq1.gif

12870_2017_1068_Article_IEq1.gif

12906_2016_1240_Article_IEq1.gif

12888_2016_877_Article_IEq11.gif

12864_2016_2871_Article_IEq10.gif

12864_2015_2192_Article_IEq13.gif

12864_2017_3503_Article_IEq1.gif

12888_2016_877_Article_IEq6.gif

12888_2016_877_Article_IEq5.gif

12864_2015_2192_Article_IEq10.gif

12864_2016_2789_Article_IEq16.gif

12888_2016_877_Article_IEq2.gif

12864_2016_3105_Article_IEq10.gif

12864_2015_2214_Article_IEq2.gif