[Retracted] Improving Recognition of Overlapping Activities with Less Interclass Variations in Smart Homes through Clustering-Based Classification

Table 11

Performance evaluation metrics on the Milan dataset with activities balancing using threefold cross-validation.

Dataset	Cross-validation	Clustering method	Classification method	Precision (%)	Recall (%)	F score [0, 1]	Accuracy (%)
Milan	Threefolds	Fuzzy C-means [35]	ANN [45]	92.20	92.50	0.92	92.20
			ET-KNN [47]	86.20	86.50	0.86	86.70
			KNN [48]	84.30	85.40	0.85	85.10
			SMO [46]	80.50	80.30	0.80	80.30
		Hierarchical [42]	ANN	89.30	89.40	0.89	89.40
			ET-KNN	85.20	85.30	0.85	85.30
			KNN	83.30	83.80	0.83	83.80
			SMO	79.40	79.10	0.79	79.50
		K-mean [36]	ANN	85.50	85.10	0.85	85.10
			ET-KNN	83.20	82.80	0.83	83.50
			KNN	79.70	80.30	0.80	80.50
			SMO	77.30	76.02	0.75	76.02
		DBSCAN [37]	ANN	83.30	83.80	0.83	83.80
			ET-KNN	80.20	80.40	0.80	80.20
			KNN	79.50	79.20	0.79	79.20
			SMO	78.02	78.20	0.78	78.30

The precision, recall, and accuracy are in percentages (%), while the range of F score is between [0-1] with 1 being the highest. The highest values are in bold.