Title: AI-Driven Adaptive Sorting Algorithms for Large-Scale Data Processing
Authors: Altaher Almoghrabi and Samy S. Abu-Naser
Volume: 9
Issue: 8
Pages: 155-165
Publication Date: 2025/08/28
Abstract:
In today's data-driven world, the efficient sorting of large-scale datasets is critical for optimizing system performance across diverse domains such as search engines, financial analysis, bioinformatics, and real-time analytics. Traditional sorting algorithms such as Quicksort, Mergesort, and Heapsort, while computationally powerful, often fail to dynamically adapt to heterogeneous data characteristics or varying computational environments. This paper proposes the integration of Artificial Intelligence (AI), particularly machine learning (ML) and reinforcement learning (RL), to design adaptive sorting systems capable of analyzing dataset features in real time and selecting the optimal sorting strategy. We introduce a hybrid framework consisting of a Data Profiler, Sorting Strategy Selector, and RL-based Optimizer. Experimental evaluations conducted on both synthetic and real-world datasets containing up to 50 million elements demonstrate significant reductions in execution time (20-40%) and improved scalability over traditional algorithms. This work contributes to the growing field of AI-based algorithmic optimization and suggests potential for integration into large-scale data processing pipelines.