Efficient query processing is a fundamental requirement in relational database management systems (RDBMS), particularly in large-scale enterprise and cloud environments. Database indexing plays a crucial role in improving query performance; however, traditional index tuning methods rely heavily on manual intervention and static heuristics, which are inefficient for dynamic workloads. This paper proposes an AI-Based Database Index Optimization Model (AIDIOM) that leverages machine learning and reinforcement learning to automatically generate, evaluate, and optimize database indexes. The system continuously monitors workload patterns, extracts query features, predicts index utility using supervised learning models, and refines decisions using reinforcement learning feedback. Experimental evaluation using standard benchmarks such as TPC-H demonstrates significant improvements in query execution time, throughput, and resource utilization compared to traditional indexing methods. The results confirm that AI-driven index optimization provides a scalable and adaptive solution for modern database systems.
Database Indexing, Machine Learning, Query Optimization, Reinforcement Learning, Relational Database, AI for Databases, Performance Optimisation.
Dr. Surender Singh. AI-Based Database Index Optimization Model for Intelligent Query Performance Enhancement in Relational Database Systems. Indian Journal of Modern Research and Reviews. 2026; 4(SP1):313-316
Download PDF
