A Discrete Wavelet Based Feature Extraction and Hybrid Classification Technique for Microarray Data Analysis

Table 6

Performance analysis of leukemia set using rbio2.2 with variations in window sizes and levels.

Classifier	Window size	Classification accuracy (%)
		Level 1	Level 2	Level 3	Level 4	Level 5	Level 6	Level 7
KNN	64	81.56	83.86	88.21	0.00	0.00	0.00	0.00
	128	81.26	82.41	87.61	83.56	0.00	0.00	0.00
	256	68.57	62.52	74.06	81.56	94.25	0.00	0.00
	512	63.97	74.91	62.22	61.97	88.21	89.36	0.00
	1024	60.22	75.51	60.17	83.01	59.37	85.31	56.47
Bayes	64	92.80	91.35	86.46	0.00	0.00	0.00	0.00
	128	81.01	85.31	90.75	86.46	0.00	0.00	0.00
	256	70.91	92.50	77.81	80.11	87.61	0.00	0.00
	512	83.01	72.86	69.17	68.62	83.86	91.35	0.00
	1024	64.27	81.86	65.97	87.91	61.12	85.61	61.37
SVM	64	91.35	87.31	92.80	0.00	0.00	0.00	0.00
	128	87.06	78.41	92.50	85.31	0.00	0.00	0.00
	256	72.91	85.86	73.46	77.51	68.62	0.00	0.00
	512	74.91	75.21	63.72	63.72	83.61	84.16	0.00
	1024	67.72	80.71	61.92	82.71	64.57	85.31	59.37
Hybrid	64	98.61	100.00	100.00	0.00	0.00	0.00	0.00
	128	100.00	90.28	98.61	97.22	0.00	0.00	0.00
	256	95.83	95.83	90.28	100.00	100.00	0.00	0.00
	512	93.06	95.83	93.06	79.17	100.00	97.22	0.00
	1024	76.39	88.89	79.17	93.06	83.33	94.44	76.39