Claim Missing Document
Check
Articles

Found 33 Documents
Search

Classification Of Outstanding Students Using Support Vector Machine (SVM) Based on Data Mining Riduan Syahri; Desi Puspita
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 9 No. 1 (2025): Issues July 2025
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v9i1.13191

Abstract

This research aims to classify outstanding students at the Pagar Alam Institute of Technology using the Support Vector Machine (SVM) algorithm based on data mining. Early identification of outstanding students is crucial for supporting potential development and institutional decision-making. Historical data from 245 students from the 2016 to 2018 cohorts were utilized, encompassing course grades and Cumulative Grade Point Average (CGPA). The research process included data preprocessing such as normalization and splitting the data into 80% training data and 20% testing data. The SVM model was implemented with a Radial Basis Function (RBF) kernel and parameters C=1.0 and gamma=0.1. Evaluation results show that the model achieved an overall accuracy of 89.80% on the testing data. The model's performance was further validated through a confusion matrix (9 True Positives, 1 False Negative) and a classification report indicating good precision and recall for both classes. Furthermore, an Area Under the Curve (AUC) value of 0.93 signifies the model's excellent discriminative ability. This study contributes by providing an effective classification tool for identifying outstanding students, which can serve as a basis for the institution to design more targeted development and recognition programs.
Penerapan Algoritma C4.5 Untuk Prediksi Anak Stunting Di Kota Pagar Alam Hakim, Revaldo Xsanal; Putrawansyah, Ferry; Syahri, Riduan
Jurnal Tekno Kompak Vol 18, No 2 (2024): AGUSTUS
Publisher : Universitas Teknokrat Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33365/jtk.v18i2.4078

Abstract

Di Pagar Alam, Prediksi dan pengukuran tingkat Stunting masih mengandalkan analisis sekunder. Kader Posyandu melibatkan diri dalam mengukur kondisi balita, dan hasilnya diserahkan kepada ahli untuk menilai apakah balita tersebut mengalami Stunting atau tidak. Tujuan dari penelitian ini adalah mengaplikasikan Algoritma C4.5 untuk melakukan Prediksi terkait kasus Stunting pada anak. Dari permasalahan yang ada diatas, maka metode yang dapat menyelesaikan permasalahan ini yaitu Algoritma C4.5 yang termasuk dalam Pohon Keputusan pada Data Mining. Proses Data Mining peneliti menggunakan salah satu metode CRIPSP-DM dan pengujian Data Mining menggunakan Confusion Matrix serta pengujian Sistem menggunakan Black Box Testing. Hasil dalam penelitian ini berupa sebuah Sistem yang menerapkan aturan dari Pohon keputusan. Sistem Prediksi status gizi Anak yang dirancang penulis layak karena dapat mengkategorikan status gizi balita. balita secara otomatis berdasarkan Zscore yang ditetapkan dan hanya terdapat selisih 11,8% dari pengujian Prediksi dataset yang sama menggunakan Rapid Miner. Sistem yang penulis rancang dapat lebih cepat dan efektif dalam memPrediksi status gizi Anak. Berdasarkan Berdasarkan data hasil uji, dapat disimpulkan bahwa akurasi Algoritma C.4.5 untuk memPrediksi anak Stunting yaitu 88,20% tergolong baik. Sedangkan pengujian Sistem menggunakan Black Box Testing memperoleh total 4.35 masuk ke kategori sangat layak. Tujuan Sistem ini adalah membantu tenaga kesehatan dalam membuat keputusan terkait Prediksi status gizi balita. Sistem Prediksi ini bermanfaat untuk mengidentifikasi balita berisiko gizi buruk sehingga tindakan pencegahan dapat dilakukan dengan lebih efektif.
Optimalisasi Kader Posyandu Kelurahan Bangun Rejo Pembuatan Produk Makanan Tambahan Seblak (Sehat, Bergizi, Lahap, Enak) Putrawansyah, Ferry; Syahri, Riduan; Padya, Inka Rizki
Jurnal Pengabdian Masyarakat (ABDIRA) Vol 5, No 4 (2025): Abdira, Oktober
Publisher : Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/abdira.v5i4.1147

Abstract

The provision of supplementary feeding at community health posts aims to support adequate nutrition for toddlers. However, limited government funding poses challenges for posyandu cadres. To overcome this, cadres conduct self-funding initiatives, even though most of them are housewives. This community service activity aims to assist cadres in developing a provision of supplementary feeding production enterprise through education, capital support, and business mentoring. The stages include preparing health media, conducting health education, forming a management team, producing provision of supplementary feeding, and evaluating business progress. The activity involved 12 cadres from Bangun Rejo Village. The results showed increased knowledge among cadres in producing nutritious, locally sourced provision of supplementary feeding, along with improvements in toddlers’ nutritional status: severe malnutrition decreased from 25% to 10%, moderate malnutrition from 75% to 50%, and good nutrition increased from 0% to 40%. This demonstrates that PMT effectively improves toddlers’ nutrition.
SISTEM PENDUKUNG KEPUTUSAN REKOMENDASI HOMESTAY DI KOTA PAGAR ALAM DENGAN METODE TOPSIS Masdalipa, Risnaini; Setiadi, Dedi; Syahri, Riduan
(JITEK)Jurnal Ilmiah Teknosains Vol 9, No 2/Nov (2023): JiTek
Publisher : Universitas PGRI Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26877/jitek.v9i2/Nov.17247

Abstract

Technological developments can be used appropriately and properly will provide benefits and facilitate human work. The decision support system is an information system that can be used to assist in decision making, in this case, namely providing homestay recommendations to tourists who will visit the city of Pagar Alam which is a tourist destination in South Sumatra. Tourists who will stay at homestays must come directly to several homestays to ask about the rental price and facilities provided by the homestay manager. Sometimes people who are looking for a homestay don't immediately match the homestay they are visiting, so they have to find another homestay that is suitable and in accordance with the available budget so that this situation is not optimal, because the process for people to find a suitable homestay can take a long time. The method used is Technique for Order of Preference by Similiarty to Ideal Solution, which is a method for solving Multi Attribute Decision Making problems. The system development uses the Rapid Application Development (RAD) method, which is a process model used in incremental software development, especially for short processing times, which consists of three stages, namely requirements planning, design workshop, implementation. From the calculation results of the TOPSIS method with four criteria determined for several alternatives, according to the selection of criteria by the user, as well as the results of ranking alternatives with the highest preference value, namely 0.5962, and kia homestay as a recommendation according to what is expected by the user.
KLASIFIKASI PENGUNJUNG WISATA DI KOTA PAGAR ALAM DENGAN MENGGUNAKAN ALGORITMA K-NEAREST NEIGHBOR (K-NN) Syahri, Riduan; Puspita, Desi
(JITEK)Jurnal Ilmiah Teknosains Vol 9, No 2/Nov (2023): JiTek
Publisher : Universitas PGRI Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26877/jitek.v9i2/Nov.17329

Abstract

The city of Pagar Alam has many beautiful tourist options, fresh and cold air, and unique culture and culinary delights. So that it becomes an area that is visited by many local and foreign tourists. Of the many visitors who come, not a few of them leave impressions in the form of reviews of the places they have visited. The purpose of this study is to determine the classification and to determine the accuracy produced by the K-Nearest Neighbor (K-NN) method. The K-Nearest Neighbor (K-NN) method is used to classify visitor data on Pagar Alam tours. Tests carried out to get good accuracy results and evaluate using a confusion matrix. This research produces a classification system that can identify and classify Pagar Alam tourism visitors using the K-Nearest Neighbor (K-NN) algorithm with the results obtained the greatest accuracy with a value of k = 3 with 99% accuracy, K0 gets 98% precision, recall 100 and a fi-score of 99%, for k1 precision 100%, the recal is 89% and the fi-score is 92%, while for K2 the precision is 100%, the recal is 100%, the f1-score is 100%.
IMPLEMENTASI MODEL TRANSFER LEARNING RESNET152V2 UNTUK KLASIFIKASI PENYAKIT DAUN PEPAYA Yohanes Riski Ramadhan; Alfis Arif; Riduan Syahri
JATI (Jurnal Mahasiswa Teknik Informatika) Vol. 10 No. 3 (2026): JATI Vol. 10 No. 3
Publisher : Institut Teknologi Nasional Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36040/jati.v10i3.18013

Abstract

Pepaya merupakan salah satu komoditas hortikultura yang memiliki nilai ekonomi penting bagi petani di Kecamatan Pajar Bulan. Namun, produktivitas pepaya sering mengalami penurunan akibat serangan berbagai penyakit pada daun. Selama ini identifikasi penyakit masih dilakukan secara visual oleh petani dan penyuluh, sehingga berpotensi menimbulkan kesalahan diagnosis. Penelitian ini bertujuan untuk mengimplementasikan model klasifikasi penyakit daun pepaya berbasis Deep Learning menggunakan pendekatan Transfer Learning dengan arsitektur ResNet152V2 serta mengintegrasikannya ke dalam aplikasi mobile berbasis Flutter. Dataset yang digunakan berjumlah 800 citra daun pepaya yang terdiri dari empat kelas, yaitu Antraknosa, Blackspot, Ringspot, dan Healthy. Data dibagi menjadi 80% untuk training, 10% untuk validasi, dan 10% untuk testing. Tahapan pengolahan data meliputi cleaning, resize citra menjadi 224×224 piksel, normalisasi, serta augmentasi data. Proses pelatihan dilakukan dalam dua tahap, yaitu freeze dan fine-tuning masing-masing selama 100 epoch dengan memanfaatkan bobot pra-latih ImageNet. Hasil evaluasi menunjukkan model mampu mencapai akurasi sebesar 98,75% dengan nilai rata-rata Precision, Recall, dan F1-Score sebesar 0,99 serta loss yang stabil tanpa indikasi overfitting yang signifikan. Model kemudian dikonversi ke format TensorFlow Lite dan berhasil diimplementasikan pada aplikasi Flutter untuk membantu proses identifikasi penyakit daun pepaya secara lebih praktis dan efisien.
Hotspot Management Training Using Mikrotik RouterBoard to Support Independent Internet Access for the Community Dedi Setiadi; Riduan Syahri; Febriansyah; Debi Gusmaliza; Risnaini Masdalipa
JURIBMAS : Jurnal Hasil Pengabdian Masyarakat Vol 5 No 1 (2026): Juli 2026
Publisher : LKP KARYA PRIMA KURSUS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62712/juribmas.v5i1.1015

Abstract

Currently, the internet has become one of the basic needs for society, one of which is in the field of education. Where many students or college students are constrained in participating in learning or searching for materials due to internet access constraints, so hotspot creation training is needed. The creation of this hotspot aims to ensure that all activities that require internet access can run smoothly. The training is carried out through community service activities as a solution to problems that occur in the community. The stages of implementing this community service consist of 4 (four) stages, namely the first stage of preparation by conducting surveys and interviews, the second stage of training implementation is carried out by delivering materials and practicing hotspot creation, the third stage is an evaluation by giving questionnaires to participants to determine the understanding of each participant, and the fourth stage is making reports and publications as outputs of community service activities. Based on the evaluation that has been carried out, it is known that the level of understanding and knowledge of the training participants has increased significantly. Before participating in the training, only 50% of participants knew about computer networks, increasing to 95%. Before participating in the training, only 5% of participants had ever created a hotspot with a Mikrotik router board, increasing to 90%. Before participating in the training, only 5% of participants were able to configure Mikrotik, increasing to 90%. Before the training, only 5% of participants were able to create a hotspot using a Mikrotik router board, increasing to 80%. Before the training, only 50% of participants understood the functions of the internet and hotspot, increasing to 95% after the training
Pelatihan Pengembangan Aplikasi Machine Learning Berbasis Streamlit untuk Mendukung Penyelesaian Tugas Akhir Mahasiswa Febriansyah; Dedi Setiadi; Riduan Syahri
JURIBMAS : Jurnal Hasil Pengabdian Masyarakat Vol 5 No 1 (2026): Juli 2026
Publisher : LKP KARYA PRIMA KURSUS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62712/juribmas.v5i1.1350

Abstract

Mahasiswa Teknik Informatika yang mengangkat topik machine learning dalam tugas akhir masih mengalami kesulitan dalam mengimplementasikan model yang telah dibangun ke dalam aplikasi yang dapat digunakan secara langsung oleh pengguna. Kegiatan pengabdian kepada masyarakat ini bertujuan untuk meningkatkan kompetensi mahasiswa dalam mengembangkan aplikasi machine learning berbasis Streamlit guna mendukung penyelesaian tugas akhir. Kegiatan diikuti oleh 35 mahasiswa tingkat akhir Program Studi Teknik Informatika Institut Teknologi Pagar Alam (ITPA) dan dilaksanakan melalui pelatihan, demonstrasi, praktik langsung, pendampingan, serta evaluasi menggunakan pre-test dan post-test. Hasil kegiatan menunjukkan peningkatan kompetensi peserta yang ditandai dengan kenaikan nilai rata-rata dari 54,3 pada pre-test menjadi 86,7 pada post-test. Selain itu, sebanyak 31 peserta (88,6%) berhasil membangun aplikasi berbasis Streamlit dan 27 peserta (77,1%) berhasil mengintegrasikan model machine learning ke dalam aplikasi yang dikembangkan. Tingkat kepuasan peserta mencapai 92,5%, menunjukkan respons yang sangat positif terhadap kegiatan. Hasil ini menunjukkan bahwa pelatihan berbasis Streamlit efektif dalam meningkatkan keterampilan implementasi machine learning serta mendukung penyelesaian tugas akhir yang lebih aplikatif.
Integration of Machine Learning and Web-Based Expert Systems for Diabetes Risk Analysis in Pagar Alam Riduan Syahri; Desi Puspita; Risnaini Masdalipa
Knowbase : International Journal of Knowledge in Database Vol. 5 No. 2 (2025): December 2025
Publisher : Universitas Islam Negeri Sjech M. Djamil Djambek Bukittinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30983/knowbase.v5i2.10268

Abstract

This study aims to develop an integrated system combining Machine Learning (ML) and a Web-Based Expert System for genomic and clinical data analysis to mitigate the rising diabetes cases in Pagar Alam City. The research adopts the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology, encompassing business understanding, data understanding, data preparation, modeling, evaluation, and deployment phases. Unlike previous studies relying on standard public datasets, this research integrates genomic profiles (TCF7L2 and KCNQ1 SNPs) alongside local clinical parameters from five sub-districts in Pagar Alam. Quantitative data from 640 samples were analyzed using the Support Vector Machine (SVM) algorithm. Evaluation results during the modeling phase show that the SVM model achieved a superior accuracy of 99.07%, demonstrating that integrating genomic data significantly enhances predictive precision. The web-based expert system implemented in the deployment phase provides personalized prevention recommendations based on individual risk profiles. This application is expected to serve as a strategic tool for the Pagar Alam government to enhance the effectiveness of prevention programs through localized and genetic-based interventions.
Integrasi Explainable AI (SHAP) Pada Model Machine Learning Untuk Analisis Faktor Penentu Kualitas Fasilitas Kesehatan Berbasis Web Riduan Syahri; Fitria Rahmadayanti; Alfis Arif
BETRIK Vol. 17 No. 02 (2026): Jurnal Ilmiah BETRIK : Besemah Teknologi Informasi dan Komputer
Publisher : PPPM Institut Teknologi Pagar Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36050/ndmny220

Abstract

The quality and equitable distribution of healthcare facilities are vital indicators of regional public service development. Although modern Machine Learning models such as Extreme Gradient Boosting (XGBoost) achieve exceptionally high predictive performance in classifying medical facility quality levels, their black-box nature often limits decision-making transparency for policymakers. This study aims to integrate an Explainable Artificial Intelligence (XAI) approach using SHapley Additive exPlanations (SHAP) into an XGBoost model developed in Google Colab and deployed as an interactive web dashboard via Streamlit. The dataset comprises aggregated BPJS Healthcare Facility data across Indonesian regencies/cities supplemented with operational indicators. The experimental results demonstrate superior modeling performance, achieving an accuracy of 95.15% and an F1-Score of 97.30%. Through SHAP Summary Plot analysis, Skor_Fasilitas_UGD and Total_Faskes were identified as the primary dominant factors driving high-quality facility classifications. The Streamlit web application successfully visualizes individual feature contributions (SHAP Waterfall Plot) in real-time, providing intuitive clinical and managerial transparency for public health planning