A fast way to find out the outliers (doufu110511@outlook.com)
data: the original points
outlier: inline-points and outline-points
result: inline_points
rank: different rank
level: the trend of level
anchor: the trend of anchor
Analysis:
Samples:
1d:
car:
guass:
circle:
spiral:
3d:
The algorithm works well in all dimentional, see more in '/result' directory !






