Resource Scheduling Based on Improved Spectral Clustering Algorithm in Edge Computing

Table 1

Performance comparison of algorithms.

Performance metrics	Algorithm
	ISCM algorithm	Traditional spectral clustering algorithm	k-means algorithm
Extendibility	High	High	High
Data types that can be processed	Numerical and non-numerical type	Numerical and non-numerical type	Numerical type
The clustering shape can be found	Arbitrary	Arbitrary	Convex and spherical type
Domain knowledge dependence on input parameter	Small	Large	Large
Sensitivity to noise	Sensitive	Sensitive	Sensitive
Sensitivity to input order	General	General	Sensitive
The ability to process high dimensional data	High	High	Lower
Time complexity	O(nck)	O(knt)	O(knt)