Online Book Recommendation System Based on Optimized Collaborative Filtering Using Ant Colony Optimization

Today, the amount of information on the Internet is growing rapidly, and people need some tools to find and access the right information. One of these tools is called a recommender system. Recommender systems help you navigate quickly and get the information you need. This paper proposes a fast and intuitive book recommendation system to help readers find the right book to read next. The overall architecture is presented with its detailed description. We used a collaborative filtering method. The idea behind the Ant Colony Algorithm is based on ant colonies, as suggested by its name. It leverages the idea that the more pheromones an ant releases, the more significant the goal is-in this case, the target being the location that is the shortest. Through the use of ant colony optimization, we aim to lower the effort needed for collaborative filtering to make good predictions, which in turn improves the algorithm's prediction accuracy. The results show that the Euclidean-based distance matrix approach can optimize EMBEDDING_SIZE well with the best value 0.3927 so that it has a lower loss value after optimization than before optimization. This means that optimization using Ant Colony Optimization is quite efficient, flexible, and beneficial, especially for Collaborative Filtering.