A Clustering Algorithm via Density Perception and Hierarchical Aggregation Based on Urban Multimodal Big Data for Identifying and Analyzing Categories of Poverty-Stricken Households in China

Table 3

The metric values on four UCI datasets.

Dataset	Metric	AHC	DBSCAN	EDBSCAN	NS-DBSCAN	ADBSCAN	HDBSCAN
Banknote	SC	0.308	0.321	0.481	0.483	0.484	0.485
	DBI	1.239	1.492	1.226	1.197	1.875	1.155
	ARI	0.796	0.753	0.865	0.912	0.987	0.991
	NMI	0.758	0.865	0.897	0.953	0.992	0.996
Planning relax	SC	0.178	0.136	0.171	0.215	0.231	0.271
	DBI	2.355	11.834	1.733	2.322	2.095	2.368
	ARI	0.612	0.537	0.712	0.739	0.865	0.898
	NMI	0.821	0.854	0.882	0.913	0.924	0.928
Parkinson	SC	0.258	0.201	0.212	0.255	0.276	0.281
	DBI	1.731	1.785	1.598	1.679	1.717	1.640
	ARI	0.635	0.689	0.728	0.825	0.877	0.969
	NMI	0.589	0.721	0.737	0.862	0.883	0.928
Codon usage	SC	0.265	0.238	0.275	0.281	0.283	0.296
	DBI	3.856	8.954	2.917	2.805	2.655	2.192
	ARI	0.712	0.884	0.865	0.928	0.944	0.939
	NMI	0.822	0.851	0.874	0.912	0.953	0.967
HCV	SC	0.328	0.337	0.416	0.397	0.419	0.511
	DBI	2.754	2.663	2.841	2.425	2.538	2.331
	ARI	0.714	0.821	0.885	0.836	0.918	0.986
	NMI	0.689	0.774	0.796	0.825	0.884	0.917