<![CDATA[The rapid growth of e-commerce platforms has led to an explosion in the number of available products, creating a problem of information overload for users. This situation makes it difficult for users to find products that match their personal preferences, thus reducing satisfaction and potential sales conversions. This research aims to develop an Artificial Intelligence (AI)-based recommendation system by implementing the Collaborative Filtering (CF) method to increase personalization. The research approach uses a quantitative descriptive method with a Waterfall-based Software Development Life Cycle (SDLC) system development model. The processed data consists of a user-product interaction matrix (ratings, purchase history) simulated from an e-commerce scenario. A user-based CF algorithm is implemented using cosine similarity calculations and weighted rating predictions. The implementation results show that the system is capable of generating relevant recommendations. In a simulation with a rating matrix (4 users, 6 products), the predicted rating for unrated items reached a value of up to 4.64, with the best recommendation being a product with high preference similarity among users. A simple evaluation yielded a Mean Absolute Error (MAE) of 1.0 on holdout data, demonstrating competitive accuracy compared to similar studies. This system has been shown to enhance the personalization of e-commerce services, potentially improving user experience and transaction volume.]]>
Copyrights © 2026
