首页> 外文会议>International conference on computer and network technology >An Improved Approach of Criminal Predictional Analysis Based on Multifactor
【24h】

An Improved Approach of Criminal Predictional Analysis Based on Multifactor

机译:基于多因素的刑事预测分析方法改进

获取原文

摘要

Geographical profile fs an investigative technique that analyzes special patterns of the related series of crime locations in order to predict the most probable position of the offender residence or determine possible locations of next crimes based on time and locations of the past crime scenes. It is part of the police behavioral science response, and has relationships with linkage analysis and psychological profiling. There is numerous location models exist in the literature. The basic fundamental problem features are still difficult to address simultaneously by using existing models when determining the anchor point. In this paper, we propose three models. The first is generated by the improved K-means algorithm which can predict the next criminal scene which has highest probability'. The second is generated by improved spatial mean and standard distance which can predict a smaller range of the next criminal scene. The third, named GM(1,1). take distance decay into account to predict the next crime place. Finally, we apply our models in the example of serial murders of Peter, and analyze the results with some other important factors.
机译:地理概况fs的一种调查技术,分析,以预测罪犯居住的最可能的位置或基于时间和过去犯罪现场的位置旁边的罪行的可能位置相关的系列作案地点的特殊模式。这是警方行为科学应对措施的一部分,与连锁分析和心理分析的关系。目前在文献中存在大量的选址模型。基本的根本问题的特征仍难以通过确定锚点在使用现有的模型同时解决。在本文中,我们提出了三种模式。第一是由改进的K-means算法,可预测下一个场景刑事具有最高概率”中产生。第二是通过改进的空间平均值和标准距离,该距离可以预测一个较小的范围内的下一个场景犯罪的生成。第三,命名为GM(1,1)。取距离衰减考虑到预测未来犯罪的地方。最后,我们适用于彼得系列谋杀的例子中,我们的模型,并分析其与其他一些重要因素的结果。

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