通信学报 ›› 2020, Vol. 41 ›› Issue (8): 130-140.doi: 10.11959/j.issn.1000-436x.2020119

• 学术论文 • 上一篇    下一篇

基于双向邻居修正的局部异常因子算法

杨晓晖,刘晓明   

  1. 河北大学网络空间安全与计算机学院,河北 保定 071002
  • 修回日期:2020-04-30 出版日期:2020-08-25 发布日期:2020-09-05
  • 作者简介:杨晓晖(1975- ),男,河北巨鹿人,博士,河北大学教授、硕士生导师,主要研究方向为分布计算、信息安全与可信计算|刘晓明(1993- ),男,河北望都人,河北大学硕士生,主要研究方向为分布式计算与信息安全
  • 基金资助:
    国家重点研发计划基金资助项目(2017YFB0802300)

Local outlier factor algorithm based on correction of bidirectional neighbor

Xiaohui YANG,Xiaoming LIU   

  1. School of Cyber Security and Computer,Hebei University,Baoding 071002,China
  • Revised:2020-04-30 Online:2020-08-25 Published:2020-09-05
  • Supported by:
    The National Key Research and Development Program of China(2017YFB0802300)

摘要:

针对现有离群点检测算法存在参数选取困难、效率差和精度低等问题,提出了基于双向邻居修正的局部异常因子算法。为了解决所提问题,首先提出了基于双向邻居的搜索算法,降低邻居搜索占用时间,然后使用双向邻居的修剪算法减少参数输入以及不必要的异常值计算。同时提出了基于双向邻居的修正因子,并利用反向邻居进一步提高计算精度。实验结果表明,所提算法减少了参数选取,提高了时间效率,同时基于双向邻居的修正因子使算法在合成数据集和UCI数据集上的准确率更高。

关键词: 异常值检测, 局部异常因子, 双向邻居, 修正因子

Abstract:

A local outlier factor algorithm based on bidirectional neighbor correction was proposed to solve the problems of existing outlier detection algorithms such as difficulty in parameter selection,poor efficiency and low accuracy.The bidirectional neighbor searching algorithm was used to reduce the neighbor search time.Then the bidirectional neighbor pruning algorithm was used to reduce the number of parameters and unnecessary calculations.And the correction factor based on bidirectional neighbors was used to improve the calculation accuracy.Experimental results show that the proposed algorithm has better performance in parameter selection and time efficiency than other outlier detection methods.The correction factor improves the accuracy of the algorithm,in the synthetic data set and UCI data set.

Key words: outlier detection, local outlier factor, bidirectional neighbor, correction factor

中图分类号: 

No Suggested Reading articles found!