Structure Optimization of Locally Linear Model Tree with Merging and Particle Swarm Optimization

Locally linear model tree algorithm is one of the useful techniques in modeling of complex nonlinear systems. One of the important features in the incremental algorithms such as LOLIMOT is the structure optimization of the model. In this paper, the merging algorithm is used as a supervisor of the original LOLIMOT to overcome suboptimal LLMs. Also, particle swarm optimization is used to obtain the optimal standard deviation of each LLM and leads to have an optimize structure. The simulation results show the effectiveness of the proposed extension of the original LOLIMOT algorithm to have a good precise with optimal number of neurons.