Hybrid Optimization Algorithm for Big Data Classification

Big data classification is a crucial machine learning task since it necessitates analyzing a large amount of data with complex features. The proposed hybrid optimization strategy for big data classification combines two well-known optimization methods such as differential evolution (DE) and particle swarm optimization (PSO). The proposed method aims to accurately identify data by optimizing the parameters of the SVM classifier. The recommended PSO-DE-SVM approach initially optimizes the SVM parameters with PSO. Then, PSO is used to discover the first set of SVM parameters. PSO's selected parameters are then improved by using DE in order to improve classification accuracy. The MNIST dataset and the CIFAR-10 dataset are used to test the proposed hybrid model. The findings demonstrate that PSO-DE-SVM performs better in terms of classification accuracy than other optimization techniques.