Implementasi Metode Content-Based Pada Sistem Rekomendasi Vitamin Dan Suplemen

Abstract

Since the Covid-19 pandemic started, many people have been advised to take

supplements or vitamins to boost their immune systems. There are so many different

types of vitamins and supplements available that it often confuses people when they

try to choose the right products for their health. In the current digital era, all

information can be accessed easily via online platforms and one of the popular

online platforms that provides a lot of information about vitamins and supplements

is halodoc.com. However, with all this information out there, users might feel

overwhelmed trying to find products that fit their needs. That's why we need an

effective recommendation system to help users find the best vitamin and supplement

products for their health and immune needs. One of the recommendation system

approaches that will be used in this research is the Content-Based Filtering

method. This method focuses on product characteristics, specifically the

composition of vitamins and supplements. This research uses uses a word weighting

method known as the Term Frequency-Inverse Document Frequency (TF-IDF)

algorithm and a similarity level calculation method called the Cosine Similarity

algorithm. The result of this research is a recommendation for the product

"Becefort," specifically the product "Bexicom," which has a Cosine Similarity value

of 0.8340. The recommendation system that has been developed is deployed on a

website for general access. With this recommendation system in place, it is hoped

that it will make it easier and save time for consumers in selecting replacement

vitamins and supplements that match the composition or content needed by the

consumers.