Fast Outlier Detection Using a Grid-Based Algorithm
Fig 1
(A) 530 data points in two dimensional space. The majority of the dataset belongs to one of the three clusters. Cluster I: Bivariate Normal distribution with mean (300,1000) with covariance 500. Cluster II: Bivariate Normal distribution with mean (2000,0) with covariance 50. Cluster III: Bivariate Normal with mean (3000,3000) with covariance 100000. The outlier point is generated by the uniform distribution. (B) LOF value of data points (MinPts = 10).