基于加权投票K-近邻法的生物医学缩略语消歧.pdfVIP

第22卷 第2期 中文信息学报 Vol-22，No．2 2008年3月 JOURNAL OF CHINESE INFORMATION PROCESSING Mar．，2008 文章编号：1003 0077(2008)02—0018—06 基于加权投票K一近邻法的生物医学缩略语消歧 于中华，陈 蓉，胡俊锋，陈源 (四川大学 计算机学院，四川，成都 610065) 摘 要：生物医学文献信息抽取对充分挖掘利用生物医学领域取得的重要成果，促进生物医学的进一步发展具有 重要意义。本文针对生物医学缩略语的分析理解问题，提出了基于加权投票K 近邻法的生物医学缩略语消歧算 法。该算法基于“One Sense PerDiscourse”假设自动生成带类标实例数据，消歧特征选用能表达文本主题的全局特 征词，分类算法采用加权投票K 近邻法。在包含177 762篇Medline摘要的真实语料上进行的实验表明，本文所 提出的算法明显优于相关工作中的算法。此外，实验还表明，对于缩略语消歧，加权投票K 近邻法与经典K一近 邻法相比，不但具有高的预测准确率，而且性能更加稳定。 关键词：计算机应用；中文信息处理；生物医学信息抽取；缩略语消歧；加权投票K 近邻法 中图分类号：TP391 文献标识码：A Disambiguating Biomedical AbbreViatiOns Based on K—Nearest Neighbor with Weighted Voting Method YU Zhong-hua，CHEN Rong，HU Jun-feng，CHEN Yuan (Department of Computer Science．Sichuan University，Chengdu，Sichuan 610065，China) Abstract：Information extraction from biomedical literature is very useful for utilizing the achievements in biomedical field and promoting further improvement of Biology and Medicine．This paper，aiming at biomedical abbreviation a— nalysis and understanding，proposes an approach for disambiguating biomedical abbreviations based on K nearest neighbor(K—NN)with weighted voting．In the approach，the samples with labels are generated automatically based on the hypothesis of “One Sense Per Discourse”．And the wordsdescrIhing the topic of a discourse are chosen as the features for abbreviation disambiguation．The classification model used in the approach is based 0n K—NN with weighted voting．The experimental results on a testing set containing 1 77 762 Medline abstracts show that the ap proach proposed in the paper can obtain higher precision than others in related work．