In the present study, we put forward a graph-based recommendation algorithm designed to provide personalized suggestions to users. By leveraging principles from graph theory, our algorithm utilizes the inherent structure for the recommendation system. We consider user preferences and item similarities to construct a graph representation of the recommendation network. Through a combination of graph traversal and similarity analysis techniques, our algorithm identifies relevant items for each user based on their preferences and the connections between items. The personalized suggestions generated by our algorithm aim to improve the overall user experience by offering tailored recommendations that align with individual interests.We gave a case study and for the understanding of our algorithm that may also help in comparing it against existing recommendation methods. This research contributes to the field of recommendation systems by employing graph theory principles to enhance the accuracy and personalization of the recommendation process. Also, we discuss graph homomorphism and graph isomorphism protocols to create encryption methods in cryptography.