Implementasi Algoritma Frequent Pattern Growth, Collaborative Filtering Dan Eclat Untuk Rekomendasi Produk Pada PT. JOENOES IKAMULYA

Abstract

This study aims to implement the Frequent Pattern Growth, Collaborative Filtering, and Eclat algorithms as a method for recommending products to consumers of PT. Joenoes Ikamulya. Companies are facing difficulties in determining the products that are in demand by consumers because of the increase in customer service. In order to increase product sales and expand market share, product recommendation systems to consumers are the right solution. The Frequent Pattern Growth method is used to find data sets that appear most frequently in large data sets. In this case, the algorithm generates a data tree structure called an FP Tree, which maps each transaction data into the path of the FP Tree. Collaborative Filtering leverages customer decision history to investigate reliable affiliates. By considering this affiliation, the resulting product recommendations are more accurate. Eclat's algorithm is used to find itemsets that appear frequently. Through a depth-first search on a database with a vertical layout, this algorithm is able to change a database with a horizontal layout to a vertical layout. This study integrates these three algorithms in a product recommendation system for PT. Joenoes Ikamulya. By implementing the Frequent Pattern Growth, Collaborative Filtering, and Eclat algorithms, it is hoped that the company can provide product suggestions or recommendations that match the interests and buying habits of other consumers, so as to increase sales and expand market share.

Keywords : Sistem rekomendasi; Fp Growth; Collaborative Filtering; Eclat