%0 Journal Article
%T 微博城市投诉文本中的地理位置实体识别<br>Recognition of geographical entity in city complaints of Micro-blog
%A 孙赫
%A 李淑琴
%A 吕学强
%A 刘克会&lt
%A br&gt
%A SUN He
%A LI Shu-qin
%A L(¨overU)Xue-qiang
%A LIU Ke-hui
%J 山东大学学报（理学版）
%D 2016
%R 10.6040/j.issn.1671-9352.1.2015.070
%X 摘要： 微博投诉文本中地理位置实体通常存在结构复杂,长度较长,描述较详细的特点。通过对投诉微博文本的分析,提出了地理位置实体自动识别的方法。该方法首先利用特征资源库对微博进行特征标注,使用条件随机场(conditional random fields, CRF)模型识别地理位置实体。其次根据微博和地理位置实体的特点,对CRF识别后的数据进行二次标注。最后利用微博规则库对识别结果进行补召,修正地理位置实体,最终实现地理位置实体的识别。实验结果表明该方法有显著效果,F值可达到85.52%。<br>Abstract: Geographical entity in city complaints of Micro-blog has usually has the characteristics of complicated structure, long length, the location of detailed description. This paper presents an automatic method to recognize geographical entities through analysis complaints of Micro-blog. First of all, the method utilizes the feature repository of Micro-blog to mark features, using the conditional random field(CRF)model to identify the geographical entities. Second, according to the characteristics of Micro-blog and geographical entity, recognized data by CRF is second marked. Third, rule bank is utilized to supplementing the recognition result and correcting geographical entities, consequently, the recognition of geographical entities are implemented. At last, Experimental results on the proposed method proved to have an F-Score of 85.52%
%K 地理位置实体识别
%K 微博规则库
%K CRF
%K 微博城市投诉文本
%K &lt
%K br&gt
%K city complaints of Micro-blog
%K rule bank of Micro-blog
%K recognition of geographical entity
%K CRF
%U http://lxbwk.njournal.sdu.edu.cn/CN/10.6040/j.issn.1671-9352.1.2015.070