Modeling neutrosophic variables based on particle swarm optimization and information theory measures for forest fires

Recently, neutrosophic systems modeling gained great attention in indeterminacy handling. Generating suitable membership, indeterminacy and non-membership functions for neutrosophic temperature variable in the forest fires data is a challenging problem. This paper presents a platform for integrating information theory measures with particle swarm optimization as a meta-heuristic algorithm to model a neutrosophic variable out of concrete data. Particle swarm optimization is an efficient population-based algorithm presented to find membership, indeterminacy and non-membership optimal partitions.The integrated hybrid model of information theory measures and particle swarm optimization is presented. The experimental comparisons are applied between the proposed technique and the Fuzzy C-Mean MATLAB tool for generating fuzzy membership functions. Extensive graphical results of the temperature variable of the forest fires data set and the conversion rates are provided to demonstrate the effectiveness of the proposed paradigm. The conversion rates of the proposed technique show that the optimal membership and non-membership functions are concluded after 10 and 17 iterations, respectively, which is a feasibly fast rate. Also, the proposed technique generated relatively similar functions’ subsets for the forest fire temperature variable, while the Fuzzy C-Mean clearly shifted the functions’ subsets during the experiments. Calculating the indeterminacy of the temperature in forest fire data will contribute to forecasting these fires accurately.