Next-Generation Intelligent Systems: Integrating Artificial Intelligence, Data Science, and Optimization
Abstract
Traditional methods view AI, data science, and optimization as distinct domains, limiting the capabilities of intelligent systems and their adaptability. A single framework is introduced in this paper that combines these three pillars based on a modular architecture with adaptive multi-objective optimization. It includes dynamic parameter tuning and workload management capabilities, ensuring optimal performance under different operating conditions. To reduce the calculation delay and improve efficiency of data retrieval, a novel query fragment caching algorithm is proposed in the large-scale pipelines. The proposed framework achieves a 23.7% improvement in accuracy, 31.2% decrease in latency and 28.5% increase in resource usage when compared with standalone implementations, respectively, in terms of accuracy, latency and resource usage, based on experimental evaluation on benchmark datasets. These results prove the feasibility of a pathway of autonomous decision making in complex real world environments with systematic integration.