Leather Product Recommendation System using Collaborative Filtering Algorithm
Azka Muhammad Ridwan, Diena Rauda Ramdania, Dian Sa’adillah Maylawati, Agung Wahana, Edi Mulyana, Muhammad Insan Al-Amin · 2022
Currently, e-commerce and marketplaces are constantly evolving to satisfy consumers’ needs more efficiently and conveniently. Technological developments make ecommerce smarter to serve users by providing recommendations according to user needs. Various types of products are traded in the marketplace, including leather products. Therefore, this study aims to build a recommendation system for leather products. By using the Collaborative Filtering algorithm, the system will provide recommendations for leather products to users based on patterns formed from the history of rating or user ratings. This research results in a web-based recommendation system to help users find leather products by implementing Collaborative Filtering. The experimental results on 50 leather products and 644 ratings given by 30 respondents showed an average value of Mean Absolute Error (MAE) from the application of Collaborative Filtering of 1,929. This MAE value indicates that Collaborative Filtering can recommend skin products well according to user expectations.