Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal INTECOMS: Journal of Information Technology and Computer Science STORAGE: Jurnal Ilmiah Teknik dan Ilmu Komputer
Marike Amelda Silvia Kondoj
Politeknik Negeri Manado
Civil Engineering, Building, Construction & Architecture Computer Science & IT Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
RANCANG BANGUNG SMART TUTOR STEM BERBASIS LEARNING ANALYTICS UNTUK MENDUKUNG SDG 4 PENDIDIKAN BERKUALITAS DI SD GMIM 23 MANADO Bertrand Ifanema Zega; Abel Enricko Dominique Sugiantho; Hendriko Mauliate Maraden Pakpahan; Christopel Hamonangan Simanjuntak; Marike Amelda Silvia Kondoj
INTECOMS: Journal of Information Technology and Computer Science Vol. 9 No. 2 (2026): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/bwf2c449

Abstract

The low proportion of elementary school students achieving minimum proficiency in STEM (Science, Technology, Engineering, and Mathematics) learning has become a significant challenge to achieving Sustainable Development Goal (SDG) 4 on Quality Education, particularly Indicator 4.1.1. PISA 2022 data placed Indonesia 70th out of 79 participating countries in mathematics performance, while the 2023 National Assessment showed that more than 60% of elementary school students had not achieved minimum numeracy competency. This study aims to design and develop a Smart Tutor STEM platform based on learning analytics using the K-Means Clustering algorithm to provide personalized adaptive learning experiences for students in grades 4–6 at SD GMIM 23 Manado. The method employed is Design Science Research (DSR), consisting of six stages: problem identification, definition of solution objectives, design and development, demonstration, evaluation, and communication. The platform is developed using React.js as the frontend, Python FastAPI as the backend, PostgreSQL as the database, and Scikit-learn for implementing K-Means Clustering. The K-Means algorithm groups students into three clusters based on five learning feature vectors: average STEM scores, task completion time, frequency of material repetition, answer accuracy level, and learning consistency. Clustering quality is validated using the Silhouette Score and Davies-Bouldin Index. Platform effectiveness is measured using pre-post tests with paired t-tests and N-Gain analysis, while usability is assessed using the System Usability Scale (SUS). The expected outcomes are: (1) a functional and adaptive Smart Tutor STEM platform with a Silhouette Score ≥ 0.5; (2) a significant increase in the proportion of students achieving the Minimum Mastery Criteria (KKM) in STEM (p-value < 0.05); (3) an SUS score ≥ 70 (Good category), indicating that the platform is easy to use; and (4) a tangible contribution to achieving SDG Indicator 4.1.1 at SD GMIM 23 Manado. Kata Kunci: Smart Tutor, STEM Education, Learning Analytics, K-Means Clustering, Design Science Research, SDG 4, Pendidikan Adaptif.  
IMPLEMENTASI ALGORITMA NAÏVE BAYES UNTUK PREDIKSI RISIKO RELAPS PADA SISTEM MONITORING PASIEN REHABILITASI NARKOBA Steve Jeremy Rarumangkay; Yonathan Kazu Datumbanua; Levi Elia Pitoy; Marike Amelda Silvia Kondoj; Franky Gerald Cliford Manoppo
STORAGE: Jurnal Ilmiah Teknik dan Ilmu Komputer Vol. 5 No. 3 (2026): Agustus
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/storage.v5i3.8209

Abstract

Penelitian ini bertujuan mengimplementasikan algoritma Gaussian Naïve Bayes pada sistem Analisis Risiko Relaps untuk memprediksi risiko relaps pasien rehabilitasi narkoba secara cepat, terstruktur, dan berbasis data. Sistem dikembangkan menggunakan Python, Streamlit, dan MySQL dengan pendekatan Research and Development. Tahapan penelitian meliputi pengumpulan data, prapemrosesan, encoding, standardisasi, SMOTE, pembagian data 80:20, pelatihan model, evaluasi, dan implementasi sistem. Variabel prediktor mencakup usia, riwayat pakai, hasil tes urine, lama pakai, status kerja, dan dukungan keluarga. Dari 87 data sampel pemeriksaan dengan 76 pasien unik, sistem mengklasifikasikan 44 pasien berisiko tinggi dan 32 pasien berisiko rendah. Evaluasi awal pada 18 data uji menghasilkan accuracy, precision, recall, dan F1-score sebesar 1,00. Meskipun hasil ini menunjukkan performa model yang sangat baik pada data uji, ukuran sampel pengujian yang masih terbatas menyebabkan hasil tersebut belum dapat digeneralisasi secara luas. Oleh karena itu, sistem masih memerlukan pengujian lebih lanjut menggunakan dataset yang lebih besar, lebih beragam, serta validasi silang untuk memastikan stabilitas dan generalisasi model sebelum diterapkan secara operasional dalam skala luas.
Co-Authors Abel Enricko Dominique Sugiantho Bertrand Ifanema Zega Franky Gerald Cliford Manoppo Hendriko Mauliate Maraden Pakpahan Levi Elia Pitoy Simanjuntak, Christopel Hamonangan Steve Jeremy Rarumangkay Yonathan Kazu Datumbanua