A New Collaborative Recommendation Approach Based on Users Clustering Using Artificial Bee Colony Algorithm

Table 1

The ratio of 𝐾-means using ABC to traditional 𝐾-means in metric 𝐷/𝐿.

	Number of instances	Number of attributes	𝑘	𝐷/𝐿
				Ratio of 𝐾-means using ABC to traditional 𝐾-means
Balance	625	4	3	0.873
Cancer	569	30	2	0.862
Cancer-Int	699	9	2	0.871
Credit	690	15	2	0.891
Dermatology	366	34	6	0.886
Diabetes	768	8	2	0.879
Ecoli	327	7	5	0.881
Glass	214	9	6	0.895
Heart	303	75	2	0.868
Horse	364	27	3	0.859
Iris	150	4	3	0.860
Thyroid	215	5	3	0.867
Wine	178	13	3	0.876

What we can know from this table is that our method is better than traditional 𝐾-means and it has more applicable cluster results.