Abstract:

Machine learning (ML) has witnessed remarkable advancements, driven by the increasing availability of large-scale data. However, applying traditional ML algorithms to such massive datasets often results in computational inefficiencies, slow training times, and memory limitations. This paper presents a novel approach to optimizing machine learning algorithms by combining data pre-processing techniques with parallel computing strategies. Specifically, we introduce an adaptive hybrid framework that leverages both dimensionality reduction and parallelization to enhance the scalability and efficiency of ML models. The framework is tested on several well-known datasets, demonstrating significant reductions in computational time while maintaining or even improving model accuracy. Our results highlight the potential of integrating optimization strategies to tackle the challenges of big data in machine learning, providing a pathway for future research and development in this area.

Description:

.