Article Per Year (5 Year)
p-Index From 2021 - 2026
3.877
P-Index
This Author published in this journals
All Journal JOIV : International Journal on Informatics Visualization Sinkron : Jurnal dan Penelitian Teknik Informatika JURNAL MEDIA INFORMATIKA BUDIDARMA Jurnal Pilar Nusa Mandiri JOURNAL OF APPLIED INFORMATICS AND COMPUTING Jurnal Infomedia CSRID (Computer Science Research and Its Development Journal) Abdimas Galuh: Jurnal Pengabdian Kepada Masyarakat MEANS (Media Informasi Analisa dan Sistem) Jurnal Teknik Informatika (JUTIF) Journal of Applied Computer Science and Technology (JACOST) SmartComp Intechno Journal : Information Technology Journal The Indonesian Journal of Computer Science SITEKNIK: Sistem Informasi, Teknik dan Teknologi Terapan
Abdulloh, Ferian Fauzi
Universitas AMIKOM Yogyakarta
Agriculture, Biological Sciences & Forestry Humanities Automotive Engineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Electrical & Electronics Engineering Engineering Law, Crime, Criminology & Criminal Justice Nursing

Published : 24 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 24 Documents
Search

 1   2   3 
Comparison Of Efficientnet And Yolov8 Algorithms In Motor Vehicle Classification Ferian Fauzi Abdulloh; Favian Afrheza Fattah; Devi Wulandari; Ali Mustopa
SITEKNIK: Sistem Informasi, Teknik dan Teknologi Terapan Vol. 2 No. 3 (2025): July
Publisher : RAM PUBLISHER

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.5281/zenodo.16561038

Abstract

The YOLOv8 accuracy curve highlights clear overfitting. As shown in the graph, the model reaches 100% training accuracy from the first epoch and remains flat, indicating it memorized the training data. However, validation accuracy lags behind, fluctuating between 90% and 92% without significant improvement. This discrepancy between training and validation performance suggests that YOLOv8 struggles to generalize to unseen data. The issue likely stems from its architecture, which is optimized for object detection tasks that prioritize object localization over feature extraction for classification. When repurposed for classification, YOLOv8 may not extract the nuanced visual patterns needed to differentiate similar classes, such as trucks and buses. Consequently, although YOLOv8 performs well on the training set, its classification accuracy in real-world scenarios is limited. Addressing this may require architectural adjustments, stronger regularization, or more diverse training data to enhance the model’s generalization for pure classification tasks.
COMPARISON OF K-NEAREST NEIGHBOR AND SUPPORT VECTOR MACHINE ALGORITHM OPTIMIZATION WITH GRID SEARCH CV ON STROKE PREDICTION Aprilliandhika, Wahyu; Abdulloh, Ferian Fauzi
Jurnal Teknik Informatika (Jutif) Vol. 5 No. 4 (2024): JUTIF Volume 5, Number 4, August 2024
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2024.5.4.1951

Abstract

Stroke ranks second as the leading cause of death globally, with disability being the primary accompanying factor. The cause of death in stroke patients is due to the lack of an optimal stroke prediction system; therefore, identifying whether a patient is experiencing a stroke or not becomes the focus of this research. Thus, the objective of this study is to compare the performance of stroke prediction using two classification models, namely K-Nearest Neighbors (KNN) and Support Vector Machine (SVM), with and without using the GridSearchCV optimization technique. In this experiment, the dataset is processed and divided into training and testing data using the SMOTE oversampling technique. Initial testing is conducted without GridSearchCV. The results of the initial testing show that the KNN model performs better than SVM, with accuracies of 91% and 83%, respectively. After optimizing parameters using GridSearchCV, both models experience a significant performance improvement. The KNN model increases accuracy to 95% with precision of 91% and recall of 98%, while the SVM model increases accuracy to 94% with precision of 90% and recall of 99%. These results indicate that using GridSearchCV to optimize parameters of KNN and SVM models can significantly enhance stroke prediction performance. There are differences in precision and recall between KNN and SVM. The KNN model tends to have higher recall, while the SVM model has higher precision, and for accuracy, the KNN algorithm outperforms SVM in stroke prediction.
Optimasi Analisis Sentimen terhadap Kinerja Direktorat Jenderal Pajak Indonesia Melalui Teknik Oversampling dan Seleksi Fitur Particle Swarm Optimization Sholihah, Nafiatun; Abdulloh, Ferian Fauzi; Rahardi, Majid
Smart Comp :Jurnalnya Orang Pintar Komputer Vol 12, No 4 (2023): Smart Comp: Jurnalnya Orang Pintar Komputer
Publisher : Politeknik Harapan Bersama

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30591/smartcomp.v12i4.5814

Abstract

Dalam domain kebijakan publik dan tata kelola pemerintahan, isu perpajakan senantiasa menjadi perhatian khusus di kalangan masyarakat. Dengan tujuan mendapatkan pemahaman yang lebih mendalam tentang pandangan publik terhadap performa Direktorat Jenderal Pajak Indonesia, penelitian ini mengadopsi pendekatan analisis sentimen, menggunakan dataset komentar yang terkumpul dari platform media sosial YouTube. Salah satu kendala signifikan yang dihadapi dalam analisis ini adalah ketidakseimbangan data sentimen komentar, dengan dominasi sentimen positif atau negatif. Dengan demikian, kami menerapkan teknik SMOTE oversampling dan Particle Swarm Optimization (PSO) sebagai strategi seleksi fitur, sebagai bagian dari upaya meningkatkan kualitas model analisis sentimen. SMOTE akan membuat data sintetis dari kelas minoritas sehingga data train akan berimbang dan tidak menghasilkan model yang mengandung bias yang disebabkan ketidak seimbangan data. Selanjutnya dilakukan pemilihan fitur yang dianggap memuat informasi penting untuk meningkatkan performa dari suatu model.Metode ini terbukti efektif, khususnya pada skenario dengan pembagian data latih sebanyak 70%. Di sini, nilai recall meningkat dari 0.47 menjadi 0.52, sebuah peningkatan yang signifikan dalam mendeteksi sentimen minoritas yang seringkali terabaikan dalam studi sejenis. Selain itu, teknik seleksi fitur menggunakan PSO, dengan menggunakan nilai F1 sebagai kriteria pbest, menghasilkan peningkatan substansial pada semua metrik evaluasi: akurasi mencapai 0.93, recall 0.63, presisi 0.70, dan F1 score 0.66. Ini menunjukkan keefektifan metode tersebut dalam memodelkan berbagai aspek sentimen terhadap perpajakan di Indonesia.
The Comparative Analysis of K-Nearest Neighbors Algorithm and Random Forest Regressor for House Price Prediction in Bandung City Ananda, Dimas Yudhistira; Abdulloh, Ferian Fauzi
Journal of Applied Informatics and Computing Vol. 10 No. 1 (2026): February 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i1.10718

Abstract

The rapid population growth and continuous urban expansion in Bandung have contributed to volatile and escalating housing prices, creating significant challenges for market transparency and affordability. This study aims to develop and evaluate machine-learning models to predict house prices in the Bandung region using a publicly available dataset consisting of 7,609 property records. Following the CRISP-DM methodology, the research includes data exploration, preprocessing (outlier handling using IQR, one-hot encoding, and feature standardization), model training, and performance evaluation. Two regression models K-Nearest Neighbors (KNN) Regressor and Random Forest (RF) Regressor—were compared through systematic hyperparameter tuning using Grid Search and Random Search techniques. The experimental results show that the Random Forest Regressor achieves the best performance with an R² score of 0.7838 and a mean absolute error (MAE) of approximately Rp 399.7 million, outperforming the optimized KNN model. Feature importance analysis also indicates that land area, building area, and location are the most influential predictors of property prices. The findings highlight the effectiveness of ensemble methods in handling complex real-estate data and demonstrate the potential of machine-learning-based predictive tools to support buyers, sellers, and policymakers in making informed and data-driven decisions in the Bandung housing market.
Co-Authors Abdul Muin Afrig Aminuddin Agun, Pasipikus Yosua Ali Mustopa Ananda, Dimas Yudhistira Anggita, Sharazita Dyah Aprilliandhika, Wahyu Arfan Yoga Aji Nugraha Asadulloh, Bima Pramudya Buana, Rifqi Genta Devi Wulandari Devi Wulandari Fauzy, Muhamad Ilham Favian Afrheza Fattah Febrilia Hayyu Pradaningrum, Febrilia Fetrus Jari Harianto Fitriyani, Nurul Khasanah Harianto, Fetrus Jari Jeki Kuswanto Joang Ipmawati Kusnawi Kusnawi Majid Rahardi Maulana Abrari, Muhammad Naufal Moch Farid Fauzi Nafiatun Sholihah Nurcholis, Muhammad Tofa Oktaviani, Inggit Dwi Pambudi, Iqbal Rilo Putra, Dhendy Mardiansyah Rahman, Rizki Rifqi Genta Buana Rismawati Nurul Ikhsani Teuku Muhammad Dista Wahyudi, Hasyim Sri Yusuf Amri Amrulloh