Internet of Things(IoT) based Data Analysis for Feature Selection by Hybrid Swarm Intelligence(SI) Algorithm

The feature selection is a pre-processing concept for big data and it is a complex to solve. The aim of feature selection is to remove irrelevant and redundant attributes of a specific dataset and to minimize the dimension of the feature set to manage the accuracy. Different types of techniques develop to classify the datasets. The proposed paper providing importance of Swarm Intelligence algorithms(SI). These are one of the flavour of Nature Inspired Computing(NIC) algorithms. The Nature Inspired computing algorithms are non-deterministic algorithms and these are inspired by nature. These NIC algorithms are a model for effectively solving real-time computation problems. NIC algorithms posses inbuilt features are collective behaviour, decentralized, social behaviour for efficiently solve the computational problems. The binary variants of meta-heuristic algorithms proposed for efficiently solving feature selection problem by hybridized SI algorithms. The aim of the research paper introduce a hybrid algorithms are combination of standard algorithms Particle Swarm Optimization (PSO) combined with Gray Wolf Optimization (GWO) to become more efficient and erased complexities to solve feature selection problem. In this hybrid approaching is mainly based on meta-heuristics swarm intelligence algorithms is to simulate social behaviour impact on local and global best position of particle swarm intelligence algorithms and hunting behaviour of gray wolf algorithm. This hybrid approach is acquire balance between various search methods to find optimum for features selection. As per available datasets and evaluate metrics to assess the quality and effectiveness of HGWOPSO.