| 金融领域的事件句抽取* |
| |
| 引用本文： | 李江龙,吕学强,周建设,刘秀磊. 金融领域的事件句抽取*[J]. 计算机应用研究, 2017, 34(10) |
| |
| 作者姓名： | 李江龙 吕学强 周建设 刘秀磊 |
| |
| 作者单位： | 北京信息科技大学,北京信息科技大学,首都师范大学,北京信息科技大学 |
| |
| 基金项目： | 2014年度国家社会科学基金委托课题“语言大数据挖掘与文化价值发现”（No：14@ZH036）;北京成像技术高精尖创新中心项目（No：BAICIT-2016003）;国家自然科学基金项目“基于本体的专利自动标引研究”（项目编号：61271304）;国家自然科学基金项目“中文专利侵权自动检测研究”（项目编号：61671070） |
| |
| 摘 要： | 事件句抽取是事件抽取中的核心环节,而在金融领域中,公司名识别则是事件句抽取中的重点和难点。从金融领域的事件句抽取出发,首先充分利用互联网搜索和上市公司名信息进行公司名识别：如果一个N元组是公司名,则进行互联网搜索,其结果中包含“公司”、“集团”等字词多,同时和已经的公司名有较高的匹配度。其次,综合考虑句子位置信息、包含公司名信息、包含领域动词信息、与标题相似度四个方面特征,构造权值表达式。最终从句子集中选出金融事件句。实验表明,公司名识别方法的正确率可达82.28%,召回率达68.93%;事件句抽取的正确率可达66.83%。
|
| 关 键 词： | 简称 公司名识别 事件句 事件抽取 |
| 收稿时间： | 2016-07-05 |
| 修稿时间： | 2017-07-02 |
Event sentence extraction in financial field |
| |
| Li Jianglong,Lv Xueqiang,Zhou Jianshe and Liu Xiulei. Event sentence extraction in financial field[J]. Application Research of Computers, 2017, 34(10) |
| |
| Authors: | Li Jianglong Lv Xueqiang Zhou Jianshe Liu Xiulei |
| |
| Affiliation: | Beijing Information Science and Technology University,Beijing Key Laboratory of Internet Culture and Digital Dissemination Research;China,Beijing Information Science and Technology University,Capital Normal University,Beijing Advanced Innovation Center for Imaging Technology,Beijing Information Science and Technology University,Beijing Key Laboratory of Internet Culture and Digital Dissemination Research |
| |
| Abstract: | Event sentence recognition is an important part of the event extraction, and in the financial field, the identification of the company"s name is an essential as well as a difficult part of the event sentence recognition. For the event sentence identification in the financial field, we first make full use of the information and Internet search information. Secondly, four factors are considered to construct the value of multi-factor expression: the position of the sentence, the information of the company name, the domain verb information and the similarity between sentence and title. The experimental results show that the correct rate of the company name recognition method is 82.28%, and the recall rate is 68.93%. And the correct rate of event sentence recognition was 66.83%. |
| |
| Keywords: | abbreviation company name identification event sentence event extraction |
|
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
|
