PiLiMoT: A Modified Combination of LoLiMoT and PLN Learning Algorithms for Local Linear Neurofuzzy Modeling

Table 1

The comparison of PiLiMoT with LoLiMoT and PLN (in same number of rules or same MSE).

Model	No. of rules (LLM)	MSE (train)	MSE (test)
LoLiMoT	9	0.0248	0.0140
PLN	4	0.0277	0.6612
PLN	5	failed	failed
PiLiMoT	7	0.0243	0.0145
PiLiMoT with Merge ability	6	0.0246	0.0145
PiLiMoT with Merge and Split by SA	5	0.0245	0.0144