Hybrid Disease Diagnosis Using Multiobjective Optimization with Evolutionary Parameter Optimization

Table 22

Performance of hybrid systems on hepatitis dataset.

Hepatitis
	Resampling	Without resampling	Resampling			Without resampling
			PSO	GSA	FA	PSO	GSA	FA
	Basic SVM		Parameter optimized SVM
PAC	89.03	85.16	89.68	89.68	89.68	87.10	87.10	87.10
Sensitivity	73.91	65.52	80.00	80.00	80.00	73.08	73.08	73.08
Specificity	91.67	89.68	91.11	91.11	91.11	89.92	89.92	89.92
F-measure	88.60	84.89	88.97	88.97	88.97	86.58	86.58	86.58
Recall	89.03	85.16	89.68	89.68	89.68	87.10	87.10	87.10
Precision	88.46	84.69	89.10	89.10	89.10	86.44	86.44	86.44
	Basic MLP		Parameter optimized MLP
PAC	90.32	81.94	92.26	92.26	92.26	87.10	85.16	83.23
Sensitivity	72.41	56.67	80.77	80.77	80.77	71.43	71.43	59.38
Specificity	94.44	88.00	94.57	94.57	94.57	90.55	87.31	89.43
F-measure	90.39	81.72	92.14	92.14	92.14	86.77	83.94	83.23
Recall	90.32	81.94	92.26	92.26	92.26	87.10	85.16	83.23
Precision	90.46	81.53	92.08	92.08	92.08	86.60	84.03	83.23