Content deleted Content added
we have to select k largest eigen vectors Tag: Reverted |
No edit summary |
||
Line 48:
# Calculate the Laplacian <math>L</math> (or the normalized Laplacian)
# Calculate the first <math>k</math> eigenvectors (the eigenvectors corresponding to the <math>k</math>
# Consider the matrix formed by the first <math>k</math> eigenvectors; the <math>l</math>-th row defines the features of graph node <math>l</math>
# Cluster the graph nodes based on these features (e.g., using [[k-means clustering]])
| |||