Personalised Recommendation Systems using RL

In recent years, recommendation systems have evolved significantly, with advancements in graph networks and reinforcement learning (RL) demonstrating considerable promise. This research work presents a novel approach that integrates graph-based representations of user-genre interactions with RL techniques to enhance the effectiveness of personalized recommendation systems. The integration of graph networks allows us to capture intricate relationships inherent in recommendation systems, such as user-genre interactions and item-item similarities. By modeling these relationships within a graph framework, our approach enables us to learn meaningful representations conducive to informed decision-making. Furthermore, the adoption of RL empowers the system to dynamically adapt and refine its recommendations over time by learning from user feedback. Through iterative exploration and exploitation, RL facilitates the optimization of the recommendation strategy, leading to enhanced user satisfaction.