A genetic algorithm based framework for software effort prediction

Journal of Software Engineering Research and Development

Table 15 Best learning schemes models performance (Continued)

DS	C	LS	S	MMRE	MdMRE	MMAR	MdMAR	MSA	MdSA	Pred25
3	4	2X5X1	0.80	0.66	0.48	2,680	876	58.16	72.81	0.43
3	5	7X4X6	0.79	0.55	0.55	2,536	1,021	60.42	72.88	0.39
3	5	7X5X6	0.78	0.56	0.56	2,559	1,027	60.06	72.74	0.38
3	6	2X1X6	0.80	0.94	0.55	2,872	1,017	55.17	70.14	0.38
3	6	2X4X6	0.81	0.94	0.55	2,862	1,040	55.33	70.22	0.38
3	6	2X5X6	0.81	0.93	0.54	2,845	990	55.60	70.88	0.39
3	6	7X2X6	0.78	0.57	0.57	2,586	1,040	59.64	72.03	0.37
3	7	2X1X1	0.78	0.66	0.50	2,716	921	57.60	72.27	0.42
3	7	2X5X1	0.80	0.66	0.48	2,680	876	58.16	72.81	0.43
3	8	1X4X6	0.79	0.55	0.55	2,547	1,023	60.25	72.77	0.39
3	8	1X5X6	0.78	0.57	0.56	2,571	1,050	59.86	72.38	0.38
3	8	2X4X6	0.81	0.94	0.55	2,864	1,029	55.29	70.57	0.39
3	9	2X4X1	0.80	0.66	0.48	2,686	879	58.07	72.75	0.43
3	9	2X5X1	0.79	0.66	0.49	2,691	897	58.00	72.58	0.43
4	1	6X5X3	0.74	0.80	0.65	2,815	1,236	56.06	65.63	0.34
4	2	6X5X3	0.74	0.79	0.68	2,801	1,251	56.28	66.05	0.34
4	2	6X1X3	0.74	0.80	0.65	2,799	1,234	56.32	66.04	0.35
4	3	7X1X6	0.77	0.60	0.56	2,596	988	59.47	71.15	0.39
4	3	7X5X6	0.76	0.60	0.58	2,624	1,026	59.04	70.50	0.38
4	4	6X1X3	0.74	0.79	0.68	2,801	1,251	56.28	66.05	0.34
4	4	6X5X3	0.74	0.80	0.65	2,799	1,234	56.32	66.04	0.35
4	5	2X2X1	0.76	0.56	0.54	2,768	984	56.79	69.85	0.41
4	6	5X2X1	0.71	0.68	0.72	3,143	1,304	50.94	62.65	0.31
3	4	2X1X1	0.78	0.66	0.50	2,716	921	57.60	72.27	0.42