Garuda - Garba Rujukan Digital

PREDIKSI HASIL PRODUKSI TANAMAN KOPI DI WILAYAH NTT DENGAN MENGGUNAKAN BACKPROPAGATION Iqbal Ulumando, Mohamad
Cangkok Vol 4 No 2 (2022): September : Jurnal Agroteknologi Pertanian dan Publikasi Riset Ilmiah
Publisher : Universitas Gajah Putih

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55542/jappri.v4i2.391

Tanaman Kopi merupakan salah satu hasil komoditas unggulan perkebunan di Indonesia. Hasil produksi kopi Indonesia mampu bersaing di pasar Internasional sebagai sumber Devisa bagi Negara. Salah satu Daerah di Indonesia penghasil kopi berkualitas terbaik adalah NTT, dimana hasil produksi kopi asal NTT sangat diminati di pasar dunia. Tujuan dari penelitian ini adalah untuk Memprediksi Hasil Produksi Kopi NTT setiap tahunnya dengan menggunakan Backpropagation. Penelitian ini menggunakan data dari Badan Pusat Statistik Provinsi Nusa Tenggara Timur (NTT) dengan mengambil data hasil produksi kopi dari tahun 2017 sampai tahun 2021 untuk memprediksi hasil produksi kopi dari tahun 2022 sampai tahun 2026. Dari hasil olah data diketahui bahwa nilai Mean Square Error (MSE) yaitu : 0,091556 dengan eppoch 1000. Maka hasilnya dapat diprediksi bahwa terjadi tingkat produksi kopi dari tahun 2022 sampai dengan tahun 2026 dengan rata-rata hasil produksi adalah sebesar 27.612,80 Ton. Penelitian ini hanya menggunakan data hasil produksi tahun sebelumnya dan belum memperhitungkan faktor lainnya yang dapat mempengaruhi hasil produksi kopi.

Prediksi Hasil Produksi Jambu Mete di Wilayah Provinsi Nusa Tenggara Timur Dengan Menggunakan Algoritma Backpropagation Mohamad Iqbal Ulumando
Jurnal Sistem Informasi dan Sistem Komputer Vol 10 No 1 (2025): Vol 10 No 1 - 2025
Publisher : STIMIK Bina Bangsa Kendari

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51717/simkom.v10i1.781

Dengan melihat penurunan hasil produksi jambu mete dalam 2 tahun terakhir di wilayah Provinsi Nusa Tenggara Timur, maka penelitian ini dilakukan untuk memprediksi tingkat produksi di masa yang akan datang menggunakan algoritma backpropagation sehingga dapat menekan jumlah penurunan produksi jambu mete dan melibatkan berbagai pendekatan yang bisa diterapkan untuk meningkatkan hasil panen dan mengatasi faktor yang menyebabkan turunnya produksi jambu mete. Metote yang digunakan untuk memprediksi adalah algoritma backpropagation, dimana data hasil produksi jambu mete diambil tahun 2013 sampai 2023 sebagai variabel dalam prediksi untuk tahun 2024 sampai 2035. Setelah dilakukan olah data, terjadi peningkatan produksi jambu mete di tahun 2030 dan 2031 dengan jumlah produksi sebesar 59695,58 Ton, sedangkan jumlah produksi terendah terjadi di tahun 2035 sebesar 31822,00 Ton dengan tingkat produksi rata-rata sebesar 48673,41 Ton. Prediksi ini menghasilkan Mean Square Error (MSE) sebesar 0,00727. Penelitian prediksi ini berupa data hasil produksi tahun sebelumnya dan belum memperhitungkan atau menggunakan faktor lainnya dalam memprediksi hasil produksi jambu mete.

PREDIKSI HARGA MOTOR BEKAS DI KOTA KUPANG MENGGUNAKAN METODE RANDOM FOREST Mohamad Iqbal Ulumando
SULIWA: Jurnal Multidisiplin Teknik, Sains, Pendidikan dan Teknologi Vol. 3 No. 2 (2026): SULIWA: Jurnal Multidisiplin Teknik, Sains, Pendidikan dan Teknologi, Juli 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/suliwa.v3i2.258

Used motorcycle prices in the market often fluctuate and are influenced by various factors such as brand, type, year of manufacture, engine capacity, mileage, vehicle condition, and tax status. In Kupang City, the determination of used motorcycle prices is generally still performed manually based on seller estimates or market conditions, which can lead to significant price differences. Therefore, a method is needed to estimate used motorcycle prices more objectively and accurately. This research seeks to develop a model for predicting used motorcycle prices employing the Random Forest algorithm. The dataset used consists of 200 used motorcycle records collected from used motorcycle sales data in Kupang City. The data undergoes a preprocessing stage before being used for model development. Furthermore, the dataset is partitioned into training data and testing to build evaluate prediction. The evaluation results show a Mean Absolute Error (MAE) of Rp.4,418,477, a Mean Squared Error (MSE) of 26,617,157,710,315, and a Root Mean SquSquared Error (RMSE) of Rp.5,159,181. The findings suggest that the Random Forest predict motorcycle prices with reasonably acceptable error rate, making it a useful approach for estimating used motorcycle prices based on vehicle attributes.

ITAF Kupang New Student Admission Prediction Using The Random Forest Method Mohamad Iqbal Ulumando; Orry Adrianus Mokola
PERFECT: Journal of Smart Algorithms Vol. 3 No. 2 (2026): PERFECT: Journal of Smart Algorithms, Article Research July 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/perfect.v3i2.284

New student admission is a crucial aspect of higher education academic planning. The Alberth Foenay Institute of Technology (ITAF) Kupang requires a data-driven approach to predict the number of new students in each study program to support more accurate decision-making. This study aims to predict the number of new student admissions at ITAF Kupang in the 2026/2027 academic year using the Random Forest method. The data used comes from historical data on new student admissions over the past five years (2021–2025) in three study programs: Informatics, Environmental Engineering, and Mechanical Engineering. The year and study program variables are used as input variables, while the number of new students is used as the output variable. The research stages include data pre-processing, transformation and encoding of categorical variables, Random Forest modeling, and model evaluation using Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). The model evaluation results show an MAE value of 9.11 and an RMSE of 10.58, indicating that the model has quite good predictive performance. The prediction results show that the number of new students in the 2026/2027 academic year is estimated to be 41 students for the Informatics Study Program, 24 students for Environmental Engineering, and 16 students for Mechanical Engineering. This research is expected to be a supporting basis for planning new student admissions at ITAF Kupang.

Prediction Of Repeating Object-Oriented Programming Course for Informatics Students at ITAF Kupang Using Extreme Gradient Boosting (XGBoost) Mohamad Iqbal Ulumando
PERFECT: Journal of Smart Algorithms Vol. 3 No. 2 (2026): PERFECT: Journal of Smart Algorithms, Article Research July 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/perfect.v3i2.285

The Object-Oriented Programming course is one of the core courses in the Informatics Study Program which has a fairly high level of difficulty so that some students have the potential to fail and have to repeat the course. This study aims to build a prediction model for students of the Informatics Study Program at ITAF Kupang who have the potential to repeat the Object-Oriented Programming course using the Extreme Gradient Boosting (XGBoost) algorithm based on student learning behavior data. The data used amounted to 60 students with variables including attendance, assignment grades, accuracy of assignment submission, discussion participation, quiz scores, practicum activities, and mid-term/final exam scores. The research stages include data collection, data preprocessing, training and testing data distribution, XGBoost model training, and model evaluation using Confusion Matrix, Accuracy, Precision, Recall, and F1-Score. The results of the study showed that the XGBoost model was able to perform good classification with an Accuracy value of 83.33%, Precision of 80.00%, Recall of 80.00%, and F1-Score of 80.00%. Feature importance analysis showed that quiz scores were the most influential factor in students' potential to repeat courses, followed by mid-term/final exam scores and assignment scores. The results of the study proved that student learning behavior data can be used to build an early warning system that helps lecturers and study programs identify at-risk students early on so that more effective academic mentoring can be provided.

Klasifikasi Resiko Drop Out Mahasiswa ITAF Kupang Menggunakan Random Forest Sebagai Sistem Peringatan Dini Mohamad Iqbal Ulumando
Jurnal Sistem Informasi dan Sistem Komputer Vol 11 No 1 (2026): Vol 11 No 1 - 2026
Publisher : STIMIK Bina Bangsa Kendari

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51717/simkom.v11i1.1255

Penelitian ini bertujuan untuk mengembangkan model klasifikasi resiko drop out mahasiswa pada Institut Teknologi Alberth Foenay (ITAF) Kupang menggunakan algoritma Random Forest. Dataset yang digunakan terdiri dari 200 mahasiswa dengan beberapa variabel akademik seperti IPK, jumlah SKS, rata-rata nilai, kehadiran, mata kuliah ulang, durasi studi, dan jumlah mata kuliah tidak lulus. Tahap preprocessing dilakukan melalui pembersihan data, imputasi nilai hilang, standarisasi tipe data, serta penanganan ketidakseimbangan kelas. Model Random Forest kemudian dilatih menggunakan data training dan dievaluasi menggunakan data testing. Hasil evaluasi menunjukkan bahwa model mampu melakukan klasifikasi risiko drop out dengan performa sangat baik, dengan akurasi 91%, precision 83%, recall 78%, F1-Score 0.80, dan ROC-AUC 0.94. Analisis feature importance menunjukkan bahwa IPK, kehadiran, dan jumlah mata kuliah tidak lulus merupakan variabel yang paling berpengaruh dalam penentuan risiko. Secara keseluruhan, penelitian ini menunjukkan bahwa Random Forest efektif digunakan sebagai sistem peringatan dini untuk mengklasifikasikan mahasiswa berpotensi drop out sehingga dapat membantu institusi dalam melakukan intervensi akademik secara lebih tepat sasaran.

RANCANG BANGUN SISTEM INFORMASI MANAJEMEN PENCARIAN RUMAH KOST BERBASIS WEB DI DESA PENFUI TIMUR KABUPATEN KUPANG Cyrillus Yuda Mardiyanto Pike; Mohamad Iqbal Ulumando; Risni Stefani
SULIWA: Jurnal Multidisiplin Teknik, Sains, Pendidikan dan Teknologi Vol. 3 No. 2 (2026): SULIWA: Jurnal Multidisiplin Teknik, Sains, Pendidikan dan Teknologi, Juli 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/suliwa.v3i2.283

Advances in information technology offer opportunities to improve the efficiency of data management and information delivery, including in the area of boarding house rentals. In Penfui Timur Village, Kupang Regency, the process of searching for and managing boarding houses is still carried out conventionally, making it less effective in providing fast and accurate information. This research aims to design and build a web-based boarding house search and management information system to facilitate access to information for prospective tenants and assist boarding house owners in managing data in a more structured manner. The method used is the waterfall software development model, which includes the stages of needs analysis, design, implementation, testing, and maintenance. The system was developed using the PHP programming language and a MySQL database and can be accessed through a web browser. The results show that the system is capable of providing boarding house information quickly, structured, and easily accessible. Furthermore, the system helps boarding house owners manage data more effectively than manual methods. Based on testing results, all system features function properly, making this system a potential solution for improving the effectiveness of boarding house information management and delivery.

Sistem Informasi Inventaris Barang Berbasis WEB Pada Institut Teknologi Alberth Foenay Kupang Merlin K. Talnoni; Mohamad Iqbal Ulumando; Risni Stefani
Jurnal Ilmu Komputer dan Informatika | E-ISSN : 3063-9026 Vol. 2 No. 4 (2026): April - Juni
Publisher : GLOBAL SCIENTS PUBLISHER

Show Abstract | Download Original | Original Source | Check in Google Scholar

Inventory management is a crucial aspect in supporting the smooth running of academic activities at universities. The Alberth Foenay Institute of Technology (ITAF) Kupang has a significant amount of assets, necessitating a structured, accurate, and easily accessible inventory system. This study aims to design and build a web-based inventory information system to improve the effectiveness and efficiency of campus inventory data management. The research method used is a qualitative method with observational data collection techniques, while the software development method applied is the Waterfall approach. The developed system provides data management features for goods, incoming goods, outgoing goods, warehouses, types of goods, units, and a dashboard as an information center. The results show that the web-based inventory system can facilitate faster, more accurate, and more integrated data collection, stock monitoring, and inventory reporting. With this system, inventory management at ITAF Kupang has become more organized, efficient, and easily accessible.

PREDIKSI PELUANG JUARA PIALA DUNIA FIFA 2026 MENGGUNAKAN ALGORITMA RANDOM FOREST DAN SIMULASI MONTE CARLO BERDASARKAN DATA STATISTIK PERFORMA TIM NASIONAL Mohamad Iqbal Ulumando
SULIWA: Jurnal Multidisiplin Teknik, Sains, Pendidikan dan Teknologi Vol. 3 No. 2 (2026): SULIWA: Jurnal Multidisiplin Teknik, Sains, Pendidikan dan Teknologi, Juli 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/suliwa.v3i2.296

The FIFA World Cup is the most prestigious international football competition, bringing together the best national teams from various countries. The 2026 FIFA World Cup introduces a new format featuring 48 teams, thereby increasing the complexity of predicting the eventual champion. This study aims to predict the winning probabilities for the 2026 FIFA World Cup by combining the Random Forest algorithm with Monte Carlo simulation, based on national team performance statistics. The dataset comprises the 48 participating teams, analyzed using eight variables: FIFA Ranking, Elo Rating, win percentage, average goals scored per match, average goals conceded per match, goal difference, squad market value, and past World Cup performance. The data underwent a preprocessing stage—specifically normalization—before being used to construct the Random Forest model. Model evaluation results yielded an accuracy of 84.50%, precision of 82.30%, recall of 81.70%, and an F1-score of 82.00%. Subsequently, a Monte Carlo simulation was conducted over 10,000 iterations to estimate the winning probability for each national team. The results indicate that Morocco has the highest probability of winning at 14.82%, followed by France (13.97%), Argentina (12.76%), Egypt (12.03%), and Brazil (10.88%). The findings demonstrate that the combination of the Random Forest algorithm and Monte Carlo simulation can provide objective, data-driven predictions regarding championship probabilities for international football competitions.

Title

Found 9 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 9 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 9 Documents
Search