Implementation Of Content-Based Filtering In Recommendation System For Sorting The Content Of Learning Materials In Islamic Financial Education Application

Abstract

Indonesia, home to the world's largest Muslim population, holds significant potential for developing a Sharia-based economy. However, the Sharia financial literacy index among Indonesian society remains low, standing at just 8.93% in 2019. This is attributed to a lack of financial management skills within the community. With rapid technological advancements, access to information, including Sharia finance, has become easier, yet it also brings about challenges in finding relevant information. To address this, a recommendation system needs to be established to offer guidance on Sharia financial matters. The application of content-based filtering, utilizing TF-IDF weighting and cosine similarity calculations, aids in providing suitable content recommendations. Upon successful implementation as a cloud web server, this system is seamlessly integrated into a mobile application. The article recommendation system achieves an average precision of 68%, recall of 100%, F1-score of 80%, and accuracy of 73%. Similarly, the video recommendation system attains a precision of 86%, recall of 100%, F1-score of 92%, and accuracy of 91%.