1-compute the airghbors of each data point.
2-compute the weihts Wi,j that best reconstruct each data point X
from its neighbor.
3-compute the vector Y best reconstructed by the weights Wi,j . [4]
And we have other dimention reduction such us :
- isomap
- Hessian Eigenmappin
- Multi-dimensional Scaling
Now I will speak for a use local tangent space for nonlinear dimention reduction . So what's local tangent ??