LOMA: A local outlier mining algorithm based on attribute relevance analysis

作者：

Highlights：

• We present a data reduction technique for local outlier detection.

• Particle Swarm Optimization algorithm is used to search the sparse subspace.

• We propose a local outlier mining method for high-dimensional and massive data set.

• Experimental results show that outlier efficiency and accuracy are both improved.

摘要

•We present a data reduction technique for local outlier detection.•Particle Swarm Optimization algorithm is used to search the sparse subspace.•We propose a local outlier mining method for high-dimensional and massive data set.•Experimental results show that outlier efficiency and accuracy are both improved.

论文关键词：Local outlier,Attribute relevance analysis,Particle swarm optimization,Sparse coefficient,Sparse factor

论文评审过程：Received 30 July 2016, Revised 27 February 2017, Accepted 5 May 2017, Available online 8 May 2017, Version of Record 15 May 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.05.009