• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Cloud Computing and Big Data >A Short Text Similarity Algorithm for Finding Similar Police 110 Incidents
【24h】

A Short Text Similarity Algorithm for Finding Similar Police 110 Incidents

机译：查找相似警察110事件的短文本相似算法

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Finding similar police 110 incidents from the incident dataset plays an important role in recognising related cases from which the investigators could find more clues and make a better decision on police deployment. We aim at finding 110 incidents with similar case features and semantic compared against a given incident. A short text similarity algorithm is presented. Our algorithm is developed from a novel semantic similarity algorithm Word Mover'd Distance(WMD). In order to emphasize the significance of case features in incident text, the method introduces the traditional term frequency-inverted document frequency(TF-IDF) as term weights to the WMD. Then the algorithm is verified on the practical dataset of public security department to find similar incidents, and experiments show that the algorithm is effective and can improve the accuracy in finding similar police incidents.
机译：从事件数据集中查找类似的110起警察事件在识别相关案件中起着重要作用，调查人员可以从这些案件中找到更多线索并更好地决定警察的部署。我们的目标是查找与给定事件相比具有相似案例特征和语义的110个事件。提出了一种短文本相似度算法。我们的算法是从一种新颖的语义相似度算法词移动距离（WMD）发展而来的。为了强调案例特征在事件文本中的重要性，该方法将传统的术语频率倒置文档频率（TF-IDF）作为WMD的术语权重。然后在公安部门的实际数据集上对该算法进行了验证，发现了类似的事件，实验表明该算法是有效的，可以提高发现类似警察事件的准确性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号