摘要
针对旅游领域,提出了一种基于隐马尔可夫模型(HMM)的旅游景点实体识别方法.该方法采用HMM学习算法,选取句子各态顺序遍历模型,结合词性特征和校正规则实现了旅游景点的自动识别.最后进行了旅游景点实体识别测试实验,结果表明所提方法取得了较好的效果.其中,开放测试识别准确率、召回率、F值分别达到了83.4%、95.7%、89.1%.
For the field of tourism, we propose a tourist attractions entity identification method based on the Hidden Markov Model （HMM）. The method use HMM learning algorithm, select the sentence order traversal of the state model, achieve the tourist attractions Automatic Identification with pos feature. We have took an experiment of tourist attractions entity identification, and the results showed that the method has good performance. In the opening test, the accuracy, recall, F-score respectively reached 83.4%, 95.7%, 89.1%.
出处
《昆明理工大学学报（理工版）》
北大核心
2009年第6期44-48,共5页
Journal of Kunming University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金项目(60663004)
教育部博士点基金项目(20050007023)
云南省中青年学术带头人后备人才基金项目(2007PY01-11)
云南省教育厅重点基金项目(07Z11139)
昆明理工大学博士基金项目(2006-12)
关键词
命名实体识别
HMM
旅游景点
named entity
Hidden Markov Model
tourist attractions
