Hybrid Global Optimization Algorithm for Feature Selection

This paper proposes Parallelized Linear Time-Variant Acceleration Coefficients and Inertial Weight of Particle Swarm Optimization algorithm (PLTVACIW-PSO). Its designed has introduced the benefits of Parallel computing into the combined power of TVAC (Time-Variant Acceleration Coefficients) and IW (Inertial Weight). Proposed algorithm has been tested against linear, non-linear, traditional, and multiswarm based optimization algorithms. An experimental study is performed in two stages to assess the proposed PLTVACIW-PSO. Phase I uses 12 recognized Standard Benchmarks methods to evaluate the comparative performance of the proposed PLTVACIW-PSO vs. IW based Particle Swarm Optimization (PSO) algorithms, TVAC based PSO algorithms, traditional PSO, Genetic algorithms (GA), Differential evolution (DE), and, finally, Flower Pollination (FP) algorithms. In phase II, the proposed PLTVACIW-PSO uses the same 12 known Benchmark functions to test its performance against the BAT (BA) and Multi-Swarm BAT algorithms. In phase III, the proposed PLTVACIW-PSO is employed to augment the feature selection problem for medical datasets. This experimental study shows that the planned PLTVACIW-PSO outpaces the performances of other comparable algorithms. Outcomes from the experiments shows that the PLTVACIW-PSO is capable of outlining a feature subset that is capable of enhancing the classification efficiency and gives the minimal subset of the core features.