Implementation of Artificial Neural Networks in an Android-Based Online Course Recommendation Application

Muhamad Syaiful Anwar (Gunadarma University)

COREAI: Jurnal Kecerdasan Buatan, Komputasi dan Teknologi Informasi
Vol 7, No 1 (2026): Sustainable Information Technology Innovation Supports a Digital-Based Smart Eco

The growing interest in online courses has caused information overload. Users now struggle to choose suitable programs. Machine Learning (ML) can address this problem through a Recommendation System Model based on an artificial neural network (ANN). This research aims to address this issue by designing an intelligent, ML-based recommendation system integrated within an Android app. The system uses a TensorFlow-built ML model to provide personalized course recommendations. The native Android application is developed with Kotlin and Jetpack Compose. Results show that the developed model provides recommendations that match user preferences and is successfully implemented in the online course recommendation application 'Rekomendasi Kursus Online.' Supporting features include Google account authentication, cloud storage, synchronization for favorite course lists, and dark mode. These features work as expected. User testing with the System Usability Scale (SUS) yielded a score of 83.17. This places the app in the Grade B category with a "Good" rating, showing the application is well-designed and suitable for use.