Spectral Clustering Algorithm Based on Improved Gaussian Kernel Function and Beetle Antennae Search with Damping Factor

Algorithm 1

NJW algorithm.

	Step 1: use the Gaussian kernel function to construct the similar matrix . .
	Step 2: degree matrix .
	Step 3: construct a normalized symmetric Laplacian matrix .
	Step 4: calculate the feature vector corresponding to the first k eigenvalues of , and construct the feature matrix .
	Step 5: normalize the feature matrix to obtain a normalized matrix , which contains n points in space reduced to k dimensions.
	Step 6: treat each row of as a point, and cluster them by K-means algorithm.