一种松耦合的生物医学命名实体识别算法  

Loose coupling algorithm for biomedical named entity recognition

作　　者：胡俊锋[1] 陈蓉[1] 陈源[1] 陈浩[1] 于中华[1] 

机构地区：[1]四川大学计算机学院,成都610064

出　　处：《计算机应用》2007年第11期2866-2869,共4页journal of Computer Applications

基　　金：国家自然科学基金资助项目(60473071);高等学校博士学科点专项科研基金项目(20020610007);四川大学计算机学院青年基金项目

摘　　要：生物医学命名实体识别(Bio-NER)是生物医学文献挖掘利用的基础工作。针对目前Bio-NER存在的困难和问题,提出了松耦合的Bio-NER算法LCA,该算法利用启发规则过滤器、词性模板匹配及改良的隐马尔科夫模型(HMM)识别生物医学命名实体。在GENIA corpus3.02语料库上进行的实验表明,LCA可以达到80%的准确率和89%的召回率,优于相关工作中的结果。The rapid development of biology and medicine in recent years leads to speedy accumulation of gigabyte biomedical information. How to use technical methods to mine and utilize the information becomes more and more important. Biomedical Named Entity Recognition （Bio-NER） is a basal work for mining and utihzing biomedical literatures. Concerning the difficulties and problems of the existing Bio-NER algorithms, a loose coupling algorithm named LCA for Bio-NER was proposed. The biomedical named entities were recognized based on heuristic rule filter, POS pattern matching pattern matching and modified Hidden Markov Model （HMM） approaches. The experimental results on GENIA corpus 3.02 show that the precision and recall of LCA are around 80% and 89% respectively, higher than the results of the related works.

关 键 词：生物医学命名实体 启发规则过滤器 词性模板匹配 词根匹配 隐马尔科夫模型 松耦合算法 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] TP391.4[自动化与计算机技术—控制科学与工程]