Sparse Matrix Storage Optimization Using Hybrid Methods
Shreya Srivastava, Sriya Nistala, Gorantla Samhitha, Nalini Sampath · 2025
Sparse matrix storage optimization is crucial in expanding the occurrences of datasets in scientific computation, machine learning, and high-dimensional applications, in which the traditional dense storage mechanism becomes computationally expensive and memory-consuming. This research presents a comprehensive analysis of storage algorithms along with metaheuristic approaches that uses evolutionary methods for optimization. The hybrid algorithm combines the Whale Optimization Algorithm (WOA), Compressed Sparse Column (CSC) with Compressed Sparse Row (CSR) format, and Dictionary of Keys (DOK) format to optimize sparse matrices more efficiently. This novel research collectively overcomes the problems that could be associated with using individual methods, raising up computational efficiency along with the efficient utilization of memory.