Article Per Year (5 Year)
p-Index From 2021 - 2026
2.494
P-Index
This Author published in this journals
All Journal International Journal of Advances in Intelligent Informatics EMITTER International Journal of Engineering Technology Jurnal Informatika Upgris JURNAL MEDIA INFORMATIKA BUDIDARMA JTERA (Jurnal Teknologi Rekayasa) InComTech: Jurnal Telekomunikasi dan Komputer Jurnal Eksplora Informatika Technomedia Journal Jurnal Tekno Insentif Jurnal Telematika Jurnal Teknologi Terapan Dinasti International Journal of Education Management and Social Science Mobile and Forensics Jurnal Teknik Industri Terintegrasi (JUTIN) Journal of Applied Data Sciences Journal of Dinda : Data Science, Information Technology, and Data Analytics JEECS (Journal of Electrical Engineering and Computer Sciences)
Syafrial Fachri Pane, Syafrial Fachri
Politeknik Pos Indonesia
Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Education Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Library & Information Science Mechanical Engineering Neuroscience

Published : 21 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 21 Documents
Search

 1   2   3 
Hybrid Multi-Objective Metaheuristic Machine Learning for Optimizing Pandemic Growth Prediction Adiwijaya, Adiwijaya; Pane, Syafrial Fachri; Sulistiyo, Mahmud Dwi; Gozali, Alfian Akbar
Journal of Applied Data Sciences Vol 6, No 4: December 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i4.981

Abstract

Pandemic and epidemic events underscore the challenges of balancing health protection, economic resilience, and mobility sustainability. Addressing these multidimensional trade-offs requires adaptive and data-driven decision-support tools. This study proposes a hybrid framework that integrates machine learning with multi-objective optimization to support evidence-based policymaking in outbreak scenarios. Six key indicators—confirmed cases, disease-related mortality, recovery count, exchange rate, stock index, and workplace mobility—were predicted using eight regression models. Among these, the XGBoost Regressor consistently achieved the highest predictive accuracy, outperforming other approaches in capturing complex temporal and socioeconomic dynamics. To enhance interpretability, we developed SHAPPI, a novel method that combines Shapley Additive Explanations (SHAP) with Permutation Importance (PI). SHAPPI generates stable and meaningful feature rankings, with immunization coverage and transit station activity identified as the most influential factors in all domains. These importance scores were subsequently embedded into the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to construct Pareto-optimal solutions. The optimization results demonstrate transparent trade-offs among health outcomes, economic fluctuations, and mobility changes, allowing policymakers to systematically evaluate competing priorities and design balanced intervention strategies. The findings confirm that the proposed framework successfully balances predictive performance, interpretability, and optimization, while providing a practical decision-support tool for epidemic management. Its generalizable design allows adaptation to diverse geographic and epidemiological contexts. In general, this research highlights the potential of hybrid machine learning and metaheuristic approaches to improve preparedness and policymaking in future health and socioeconomic crises.
Komparasi Model Klasifikasi Naïve Bayes Dan C4.5 Pada Data Prestasi Kerja PNS Vegita, Yola; Prianto, Cahyo; Pane, Syafrial Fachri
Jurnal Informatika UPGRIS Vol 8, No 2: Desember 2022
Publisher : Universitas PGRI Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26877/jiu.v8i2.13205

Abstract

— Salah satu bagian yang terpenting untuk mencapai keberhasilan dalam kemajuan suatu organisasi adalah SDM atau sumber daya manusia. Pegawai yang tidak menuntaskan pekerjaannya, maka target organisasi tidak akan tercapai. Berdasarkan hal tersebut, apabila seorang pegawai tidak maksimal dan tidak dapat bekerja dengan baik, pastinya akan mempengaruhi perkembangan dan kemajuan dari perusahaan atau organisasi. Untuk melakukan evaluasi  kinerja PNS Dinas Perhubungan Provinsi Jawa Barat dengan memanfaatkan hasil penilaian prestasi kerja, yang mana data yang digunakan adalah penilaian pada tahun 2020. Banyaknya pegawai membuat penilaian Prestasi Kerja menjadi sulit dan tidak dipungkiri penilaian juga terkadang dilakukan tidak objektif. Untuk melakukan suatu penilaian kerja dapat menggunakan metode pendukung, salah satunya dengan melakukan klasifikasi data pegawai dengan data mining. Penelitian ini membandingkan algoritma performance algoritma Naïve Bayes dan C4.5 dengan mengevaluasi hasil pemodelan dengan Confusion Matrix dan Classification Report. Hasilnya, C4.5 memiliki akurasi 99.12% sedangkan Naïve Bayes hanya 83%.
Penentuan rute terpendek antara dua titik di gudang menggunakan Dijkstra’s Algorithm dan Microsoft Excel Sanggala, Ekra; Pane, Syafrial Fachri; Habibi, Roni
Jurnal Teknik Industri Terintegrasi (JUTIN) Vol. 8 No. 1 (2025): January
Publisher : LPPM Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/jutin.v8i1.39060

Abstract

Sebuah gudang merupakan suatu faktor penting dalam logistik dan mempunyai peran vital dalam mengontrol dan mengurangi biaya logistik. Secara umum operasional pada gudang terdiri dari lima fungsi dasar, yaitu: receiving, sorting, storing, order picking dan delivering. Kecepatan pada order picking merupakan faktor penting untuk kepuasan pelanggan. Maka mempersingkat waktu order picking merupakan hal yang penting. Order Picking yang paling sederhana adalah saat produk yang dibutuhkan pelanggan hanya terletak pada satu rak saja, sehingga picker hanya perlu bergera dari titik awal menuju ke titik rak dimana produk berada. Permasalahan penentuan rute terpendek antara dua titik dapat didefinisikan sebagai Shortest Path Problem. Dijkstra’s Algorithm merupakan algoritma yang paling populer dalam menyelesaikan Shortest Path Problem. Untuk menyelesaikan Shortest Path Problem dengan Dijkstra’s Algorithm diperlukan sebuah tool yang dapat membantu menyelesaikan perhitungannya. Microsoft Excel merupakan salah satu tool yang sangat populer dan mudah digunakan untuk menyelesaikan berbagai perhitungan. Dengan mengkombinasikan berbagai formula yang terdapat pada Microsoft Excel terbukti bahwa perhitungan Dijkstra’s Algorithm untuk menyelesaikan Shortest Path Problem dapat dilakukan dengan baik.
PSO-Enhanced ensemble techniques for pandemic prediction and feature importance analysis Pane, Syafrial Fachri; Sulistiyo, Mahmud Dwi; Gozali, Alfian Akbar; Adiwijaya, Adiwijaya
International Journal of Advances in Intelligent Informatics Vol 11, No 4 (2025): November 2025
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v11i4.2091

Abstract

During the pandemic crisis that hit after 2020, Indonesia, like many other countries, faced tremendous challenges in areas such as health, economy, and mobility. An in-depth understanding of the dynamics and changes in these areas is essential to address the impacts of the pandemic. This research is an attempt to deeply analyze the impact of the pandemic and the most effective forecasting methods based on data and phenomena. Indonesia, with its growing economy and constantly adapting health system, faces conventional economic impacts, while its health system response tries to keep up with urgent needs driven by the spread of the virus. In the context of mobility, changes in how people move and interact significantly affect virus transmission. Modeling a pandemic event with all its complexities is not an easy task. Even more so, in finding the right method for prediction, ensemble techniques such as stacking and regression voting are emerging as promising approaches. However, deep learning and particle swarm optimization (PSO) techniques offer new innovations. The results of this study show that the ensemble vote provides the best performance in predicting confirmed positive cases and mortality based on factors of health, economic and population mobility in Indonesia. Through feature importance analysis using MDI and Tree SHAP, we conclude that factors such as active cases, the number of vaccinations, and economic indicators, such as close IDR and close IHSG, have a significant influence on the growth of confirmed positive cases. Meanwhile, recovery factors and vaccination number play an important role in the growth of the number of death cases. This study confirms that a multivariate approach that considers health, economy and mobility is the key to understanding and responding more effectively to the pandemic in Indonesia.
Klasifikasi Kanker Kulit menggunakan Custom CNN dengan SMOTE-Tomek dan Optimizer Nadam Talan, Maylinda Christy Yosefina; Pane, Syafrial Fachri; Fathonah, Rd. Nuraini Siti
Jurnal Tekno Insentif Vol 19 No 2 (2025): Jurnal Tekno Insentif
Publisher : Lembaga Layanan Pendidikan Tinggi Wilayah IV

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36787/jti.v19i2.2020

Abstract

Abstrak Deteksi dini kanker kulit sangat penting untuk meningkatkan harapan hidup, namun diagnosis konvensional seringkali subjektif. Berbeda dari penelitian sebelumnya, penelitian ini mengusulkan sebuah konfigurasi optimal yang menggabungkan tiga komponen kunci secara simultan: arsitektur Custom Convolutional Neural Network (CNN) yang ringan, penanganan ketidakseimbangan data menggunakan SMOTE-Tomek, dan optimisasi pelatihan dengan optimizer Nadam. Pendekatan terintegrasi yang dievaluasi pada dataset HAM10000 ini terbukti mampu mencapai efisiensi komputasi dan akurasi yang tinggi. Hasil eksperimen menunjukkan model mencapai akurasi validasi hingga 95.63% dan nilai F1-score ≥0.90, bahkan pada kelas minoritas seperti melanoma. Model ini juga berhasil diimplementasikan dalam aplikasi web dengan confidence score di atas 90%, membuktikan bahwa pendekatan yang diusulkan mampu memberikan solusi diagnosis yang objektif dan terukur untuk klasifikasi otomatis kanker kulit. Kata kunci: Klasifikasi Kanker Kulit; Citra Dermoskopi; Custom CNN; SMOTE-Tomek; Optimizer Nadam. Abstract Early detection of skin cancer is critical to improving survival rates, yet conventional diagnosis is often subjective. This study develops an objective dermoscopic image classification system using deep learning. The proposed model utilizes a lightweight Custom Convolutional Neural Network (CNN) architecture, combined with the SMOTE-Tomek method to handle data imbalance in the HAM10000 dataset. The training process was optimized using the Nadam optimizer with a 90:10 data split and 64x64 pixel image inputs. Experimental results show the model achieved a validation accuracy of up to 95.63% and an F1-score ≥0.90, even on minority classes like melanoma. The model, successfully implemented in a web application with confidence scores above 90%, proves to be an effective solution for automatic skin cancer classification. Keywords: Skin Cancer Classification; Dermoscopic Images; Custom CNN; SMOTE-Tomek; Nadam Optimizer. Kata kunci: Klasifikasi Kanker Kulit; Citra Dermoskopi; Custom CNN; SMOTE-Tomek; Optimizer Nadam.
Penerapan SVM dan Regresi untuk Prediksi Intensitas Sentimen Pemilu Presiden Indonesia Rionald, Valen; Pane, Syafrial Fachri; Helmi Setyawan, Muhammad Yusril
InComTech : Jurnal Telekomunikasi dan Komputer Vol 15, No 3 (2025)
Publisher : Department of Electrical Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22441/incomtech.v15i3.28525

Abstract

Dalam konteks pemilihan umum presiden Indonesia, analisis sentimen publik melalui media sosial merupakan alat yang penting untuk memahami persepsi dan reaksi masyarakat terhadap calon presiden dan kebijakan mereka. Studi ini mengembangkan model hybrid yang mengintegrasikan Support Vector Machine (SVM) dan Ridge Regression, menggunakan library BERT untuk memprediksi intensitas sentimen dari data Twitter. Pendekatan ini dirancang untuk mengatasi tantangan variabilitas ekspresi dan ambiguitas bahasa, yang sering kali mempersulit interpretasi data sentimen dengan tepat. Penelitian ini menggunakan teknik preprocessing yang komprehensif, termasuk pembersihan teks dan normalisasi data, serta penerapan teknik Synthetic Minority Over-sampling Technique (SMOTE) untuk menangani ketidakseimbangan kelas dalam dataset. Hasil dari penelitian ini menunjukkan bahwa model hybrid dapat mencapai tingkat akurasi, presisi, recall, dan F1-Score yang tinggi dengan tiga rasio yang berbeda, menegaskan keefektifan model dalam mengklasifikasikan dan mengukur intensitas sentimen. Temuan menunjukkan bahwa kombinasi SVM dan regresi, didukung dengan analisis BERT, efektif dalam mengklasifikasikan dan mengukur intensitas sentimen secara akurat. Hasil intensitas yang dijelaskan pada gambar 11 untuk kandidat Anies Baswedan mayoritas sentimen adalah netral sebesar 53.1%. Selanjutnya, pada gambar 12 untuk kandidat Prabowo Subianto netral sebesar 63.5% dan gambar 13 untuk kandidat Ganjar Pranowo dengan 62.9%.
Model Prediktif Indeks Kebahagiaan Berbasis Gradient Boosting Regressor dengan Optimalisasi Seleksi Fitur dan Implementasi Web Ferdinan, Dani; Harani, Nisa Hanum; Pane, Syafrial Fachri
JTERA (Jurnal Teknologi Rekayasa) Vol 10, No 2: Desember 2025
Publisher : Politeknik Sukabumi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31544/jtera.v10.i2.2025.59-68

Abstract

Penelitian ini menghadapi tantangan dalam memodelkan Indeks Kebahagiaan 2021 dari Badan Pusat Statistik (BPS) yang memiliki dimensi fitur sangat tinggi dan potensi redundansi, yang dapat menurunkan akurasi dan interpretabilitas model. Tujuan utama penelitian ini adalah untuk mengidentifikasi fitur-fitur paling berpengaruh dalam data tersebut untuk meningkatkan akurasi, efisiensi komputasi, dan transparansi model prediksi berbasis pohon keputusan. Metodologi mencakup pra-pemrosesan data dengan imputasi modus, transformasi Yeo-Johnson, dan Robust Scaler. Tiga algoritma regresi diuji: Decision Tree, Random Forest, dan Gradient Boosting Regressor, yang dioptimalkan menggunakan Particle Swarm Optimization (PSO). Model terbaik dievaluasi menggunakan metrik R², MSE, RMSE, dan MAE serta dianalisis lebih lanjut menggunakan SHAP untuk interpretasi. Hasil menunjukkan bahwa Gradient Boosting Regressor adalah model paling unggul dengan nilai R² sebesar 0,696 saat menggunakan 20 fitur terseleksi. Selain itu, sebagai bentuk implementasi praktis, model diimplementasikan ke dalam sebuah aplikasi web interaktif berbasis Flask yang memungkinkan pengguna memasukkan data melalui antarmuka kuisioner dan menerima prediksi indeks kebahagiaan secara real-time. Integrasi ini menjembatani hasil riset dengan pemanfaatan nyata oleh pengguna akhir.
Enhancing OCR Accuracy on Indonesian ID Cards Using Dual-Pipeline Tesseract and Post-Processing Reksiyano, Rendy Dwi; Pane, Syafrial Fachri; Awangga, Rolly Maulana
JEECS (Journal of Electrical Engineering and Computer Sciences) Vol. 10 No. 2 (2025): JEECS (Journal of Electrical Engineering and Computer Sciences)
Publisher : Fakultas Teknik Universitas Bhayangkara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54732/jeecs.v10i2.3

Abstract

Manual transcription of data from Indonesian identity cards (KTP) remains prevalent in public institutions, often resulting in inefficiencies and human errors that compromise data accuracy. While Optical Character Recognition (OCR) technologies such as Tesseract have been widely adopted. However, the performance on KTP images is still inconsistent due to non-uniform layouts, low contrast, and background noise. This study proposes a dual-pipeline OCR framework designed to enhance the recognition accuracy of Indonesian KTPs under real-world conditions. First, the pipeline performs static region segmentation based on predefined Regions of Interest (ROI), then uses dynamic keyword heuristics to locate text adaptively across varying layouts. The outputs of both pipelines are merged through a voting and regex-based post-processing mechanism, which includes character normalization and field validation using predefined dictionaries. Experiments were conducted on 78 annotated KTP samples with diverse resolutions and quality of images. Evaluation using Character Error Rate (CER), Word Error Rate (WER), and field-level accuracy metrics resulted in an average CER of 69.82%, WER of 80.20%, and character-level accuracy of 30.18%. Despite moderate performance in free-text areas such as address or occupation, structured fields achieved higher accuracy above 60%. The method runs efficiently in a CPU-only environment without requiring large annotated datasets, demonstrating its suitability for low-resource OCR deployment. Compared to conventional single-pipeline approaches, the proposed framework improves robustness across heterogeneous document layouts and illumination conditions. These findings highlight the potential of lightweight, rule-based OCR systems for practical e-KYC digitization and form a foundation for integrating deep-learning-based layout detection in future research.
Predictive Modeling of Delivery Delays in Transportation Using Machine Learning: A Comparative Study of Service Types Purnomo, Agus; Gia Ginasta, Nava; Syafrianita, Syafrianita; Pane, Syafrial Fachri
Dinasti International Journal of Education Management and Social Science Vol. 7 No. 2 (2025): Dinasti International Journal of Education Management And Social Science (Decem
Publisher : Dinasti Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38035/dijemss.v7i2.5736

Abstract

Traditional predictive models such as linear regression often struggle to capture the nonlinear interactions among operational factors that cause delivery delays in multi-category courier services. This study addresses that gap by developing and comparing machine learning (ML) algorithms to predict delivery delays across different service types at PT Pos Indonesia. The primary objective is to identify the most accurate predictive model and the dominant variables influencing delays across high-speed (Same Day, Next Day) and economical delivery services. A quantitative experimental design was employed using operational data from PT Pos Indonesia, consisting of 10,999 records and 12 variables. Three ML algorithms Logistic Regression, Random Forest, and XGBoost were trained and evaluated using standardized preprocessing, feature encoding, and stratified data splitting. Results show that Random Forest and XGBoost outperform Logistic Regression, each achieving approximately 65% accuracy with an AUC of 0.73, indicating moderate yet consistent predictive capabilities. Feature importance analysis reveals that Discount_offered, Weight_in_gms, and Prior_purchases are the most influential predictors of delivery timeliness.This study provides theoretical and practical contributions by introducing the first comparative ML framework for delay prediction in a national logistics context. The findings offer actionable insights for optimizing scheduling, load balancing, and promotional strategies, while advancing the integration of AI-based predictive analytics within postal logistics operations.
Collaboration FMADM And K-Means Clustering To Determine The Activity Proposal In Operational Management Activity Awangga, Rolly Maulana; Pane, Syafrial Fachri; Tunnisa, Khaera
EMITTER International Journal of Engineering Technology Vol 7 No 1 (2019)
Publisher : Politeknik Elektronika Negeri Surabaya (PENS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (209.522 KB) | DOI: 10.24003/emitter.v7i1.317

Abstract

Indonesian government agencies under the Ministry of Energy and Mineral Resources still use manual methods in determining and selecting proposals for operational activities to be carried out. This study uses the Decision Support System (DSS) method, namely Fuzzy Multiple Attribute Decision Decision (Fmadm) and K-Means Clustering method in managing Operational Plan activities. Fmadm to select the best alternative from a number of alternatives, alternatives from this study proposed activity proposals, then ranking to determine the optimal alternative. The K-Means Clustering Method to obtain cluster values for alternatives on the criteria for activity dates, types of activities, and activity ceilings. The last iteration of the Euclidian distance calculation data on k-means shows that alternatives that have the smallest centroid value are important proposal criteria and the largest centroid value is an insignificant proposal criteria. The results of the collaboration of the Fmadm and K-Means Clustering methods show the optimal ranking of activities (proposal activities) and the centroid value of each alternative.
Co-Authors Adiwijaya Agus Purnomo Alfian Akbar Gozali Ekra Sanggala Faisal Abdullah Fajri, Ravi Rahmatul Fathonah, Rd. Nuraini Siti Fatonah, Nuraini Siti Ferdinan, Dani Gia Ginasta, Nava Habibi, Roni Haq, Dzikri Izzatul Harani, Nisa Hanum Helmi Setyawan, Muhammad Yusril Khaera Tunnisa, Khaera Kurniawan, Alit Fajar Kurniawan, Cahya Mahmud Dwi Sulistiyo Poetra, Chandra Kirana Prianto, Cahyo Ramadhan, Bachtiar Reksiyano, Rendy Dwi Rionald, Valen Rolly Maulana Awangga, Rolly Maulana Siregar, M. Amran Hakim Talan, Maylinda Christy Yosefina Vegita, Yola Wijayanti, Diana Asri Zain, Rofi Nafiis