QL-CBR Hybrid Approach for Adapting Context-Aware Services

A context-aware service in a smart environment aims to supply services according user situational information, which changes dynamically. Most existing systems provide based on supervised algorithms. Reinforcement algorithms are another type of machine-learning algorithm that have been shown be useful dynamic environments through trial-and-error interactions. They also the ability build excellent self-adaptive systems. In this study, we aim incorporate reinforcement (Q-learning) into system relevant user’s context. To accelerate convergence learning (RL) and correct real situations, propose combination Q-learning case-based reasoning (CBR) We then analyze how incorporation CBR enables become more efficient adapt changing by continuously producing suitable services. Simulation results demonstrate effectiveness proposed approach compared traditional approach.