<![CDATA[This research aims to design and develop a personalized Adaptive Exercise and Nutrition Recommendation System, using a Reinforcement Learning (RL) approach, to overcome the limitations of conventional fitness programs that are generic and static. Unhealthy lifestyle trends have increased non-communicable diseases, but available wellness solutions fail to adapt in real-time to users' dynamic conditions—such as fatigue levels, weight changes, or nutritional responses—making them a major factor in consistent failure. This research adopts the Design Science Research Methodology (DSRM) methodology. The main problem is solved through designing an intelligent system architecture that is capable of managing and processing dynamic user data. The Deep Q-Network (DQN) algorithm is implemented effectively to produce optimal and personalized training and nutrition program recommendations on an ongoing basis. The system is trained and evaluated using synthetic data to enable simulation of complex feedback scenarios and address sensitive data privacy issues. The research output is a functional prototype (Proof-of-Concept) which will be analyzed and evaluated for its performance comparatively with conventional fitness programs to measure the level of program optimization and user consistency.]]>
Copyrights © 2026
