<![CDATA[Extracurricular activities play an important role in developing students' creativity. However, the problem that is often experienced by students in determining the choice of extracurricular activities is choosing the right type of activity and in line with the interests and talents of students. This study aims to test and compare the performance of the Naïve Bayes-based Content-based Filtering and Collaborative Filtering models in recommending appropriate extracurricular activities for students. Testing of each model is done by dividing the training data and test data in a ratio of 80% and 20%. The training process uses the RecommenderNET Library. The accuracy of the Contend-based Filtering model was tested using Naïve Bayes of the Multinomial type, while the Collaborative Filtering model used the Gaussian type of Nave Bayes. The test results of the Naïve Bayes model for Content-based Filtering show an accuracy rate of 74%, while Collaborative Filtering obtains 56%.Keywords: Recommendation System; Naïve Bayes; Library RecommenderNET Abstrak. Kegiatan ekstrakurikuler memegang peran penting dalam mengembangkan kreativitas siswa. Namun demikian, permasalahan yang sering dialami oleh siswa dalam menentukan pilihan kegiatan ekstrakurikuler adalah memilih jenis kegiatan yang tepat dan sejalan dengan minat dan bakat siswa. Penelitian ini bertujuan untuk menguji dan membandingkan kinerja model Content-based Filtering dan Collaborative Filtering berbasis Naïve Bayes dalam merekomendasikan kegiatan Ekstrakurikuler yang tepat bagi siswa. Pengujian masing-masing model dilakukan dengan membagi data latih dan data uji dalam perbandingan 80% dan 20%. Proses pelatihan menggunakan Library RecommenderNET. Akurasi model Contend-based Filtering diuji menggunakan Naïve Bayes jenis Multinomial, sedangkan model Collaborative Filtering menggunakan Naïve Bayes jenis Gaussian. Hasil uji model Naïve Bayes untuk Content-based Filtering menunjukkan tingkat akurasi 74%, sedangkan Collaborative Filtering memperoleh 56%.Kata kunci: Sistem Rekomendasi; Naïve Bayes; Library RecommenderNET]]>
Copyrights © 2022
