Whale Optimization Algorithm: An Innovative Approach for Enhancing User Preference Predictions in Recommender Systems

Recommender systems have become essential to various online platforms such as Netflix, Amazon, and social media, predicting user preferences for items such as movies, products, or content. Therefore, recommender systems can be used in machine learning to analyze user behavior and preferences. This study proposes enhancing these systems through the Whale Optimization Algorithm (WOA), a nature-inspired method that mirrors the hunting behavior of humpback whales. The WOA operates in three phases: first, identifying and prioritizing items based on user interactions; second, refining recommendations by modeling user item relationships through a spiral equation; and third, exploring new items to broaden recommendation diversity. By integrating WOA, this approach aims to improve prediction accuracy and user satisfaction, offering a scalable solution for dynamic, user-centric recommendation environments.

Keywords:

Recommender system , practical applications , net-bubble attack , Whale optimization algorithm (WOA)