Improvement of K-Means Algorithm and Its Application in Air Passenger Grouping

Table 3

Clustering results of k-means and k-means2o on real-world datasets.

Dataset	ACC		ARI		NMI		FMI
	k-	k-2o	k-	k-2o	k-	k-2o	k-	k-2o
Breast-cancer	0.8541	0.8910	0.4914	0.6062	0.4647	0.5276	0.7915	0.8286
Banknote	0.6122	0.5954	0.0485	0.0356	0.0303	0.0239	0.5517	0.5231
Bupa	0.8550	0.5768	0.0000	0.0058	0.0000	0.0112	0.6192	0.5136
Compound	0.6566	0.6365	0.5378	0.5043	0.7191	0.6557	0.6422	0.6181
Ct	0.8235	0.8325	0.4160	0.4399	0.3296	0.3485	0.7078	0.7199
Hayes–Roth	0.4393	0.4469	0.0202	0.0226	0.0287	0.0317	0.3501	0.3519
Iris	0.8933	0.8933	0.7302	0.7302	0.7581	0.7581	0.8208	0.8208
Libras	0.4277	0.4416	0.3199	0.2760	0.6066	0.5716	0.3734	0.3389
Parkinsons	0.7230	0.6307	0.0000	0.0625	0.0000	0.0493	0.7444	0.5889
Penbased	0.7674	0.6035	0.5992	0.4907	0.6927	0.6723	0.6412	0.5582
Vowel	0.3636	0.3645	0.2028	0.2204	0.4141	0.4337	0.2789	0.2868
Waveform21	0.5016	0.5018	0.2536	0.2547	0.3622	0.3654	0.5039	0.5047
Waveform40	0.5146	0.5160	0.2516	0.2530	0.5023	0.5035	0.3605	0.3632
Wdbc	0.8541	0.8910	0.4914	0.6062	0.4647	0.5276	0.7915	0.8286
Wine	0.7022	0.7022	0.3711	0.3675	0.4287	0.4164	0.5835	0.5809
Maximum	7	10	6	10	6	10	8	8

The best clustering performance evaluation values are shown in bold.