Linearization of Isomap
Jing Wang · 2007
The problem of dimensionality reduction arises in many fields of information processing. In this paper, we propose a novel linear dimensionality reduction algorithm called Linear Isomap (Lisomap). It preserves the geodesic distances in the low-dimensional space which is linearly mapped from the high-dimensional space. Numerical examples are given to show the improvement and efficiency of the proposed algorithm.