Novel Hybrid Squirrel-Wing Dragon Search Optimization for Big Data Analysis and Computerized Data Processing

Abstract

Introduction: Big data includes many different types of data, data velocity, and even actual and meaningful data. In addition, it is more advantageous for data management than more conventional data-processing methods.
Objective: This study suggests a useful framework based on the Hybrid Squirrel-Wing Dragon Search Optimization (HSWDSO) for preprocessing and data classification in a large data environment. The volume and variety of data have been used to mark weights.
Methods: Weights assigned on size, content, and keywords have been used to perform data preparation. Subsequently, HSWDSO are applied for both minimization and maximization scenarios, taking into consideration other computational aspects such as uniform distribution, epochs, random initialization, iterations, and time limitations.
Results: By using an impartial random process, the weight assignments were completed automatically. For the separated data, it has been done on a 0–1 scale. Prioritization and ranking have been accomplished using the Analytic Hierarchy Process (AHP) approach.
Conclusion: DOA-AHP yielded an overall average classification accuracy of 98 %, while SSO-AHP yielded an accuracy of 95 %. When compared to SSO-AHP, the DOA-AHP technique performs better overall.