A Hybrid Approach for Modular Neural Network Design Using Intercriteria Analysis and Intuitionistic Fuzzy Logic

Table 8

Results from the simulations.

Number	Description	Number of the inputs				Number of weight coefficients
1	Module 1—2; Module 2—1, 3, 4	4	0.5725	0.0084	0.5726	175
2	Module 1—2; Module 2—3, 4 (without 1)	3	0.5725	0.0321	0.5749	140
3	Module 1—2; Module 2—1, 4 (without 3)	3	0.5725	0.6667	1.2898	140
4	Module 1—2; Module 2—1, 3 (without 4)	3	0.5725	0.0527	0.5793	140
5	Module 1—2; Module 2—4 (without 1, 3)	2	0.5725	0.0631	0.5846	105
6	Module 1—2; Module 2—3 (without 1, 4)	2	0.5725	0.0807	0.5907	105
7	Module 1—2; Module 2—1 (without 3, 4)	2	0.5725	0.6667	1.2898	105