Article Per Year (5 Year)
p-Index From 2020 - 2025
0.408
P-Index
This Author published in this journals
All Journal JOURNAL OF APPLIED INFORMATICS AND COMPUTING Building of Informatics, Technology and Science
Saefulloh, Nandang
Unknown Affiliation
Computer Science & IT

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

Found 2 Documents
Search

 1 
Implementasi Algoritma Convolutional Neural Network (CNN) Untuk Klasifikasi Kecacatan Pada Proses Welding di Perusahaan Manufacturing Saefulloh, Nandang; Indra, Jamaludin; Rahmat, Rahmat; Juwita, Ayu Ratna
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5321

Abstract

Manufacturing industry has become one of the largest sectors in Indonesia, driven by increasing demand from the public. A primary concern to meet both local and international market needs is product quality. In ensuring high-quality standards, production processes require strict quality control. One common issue in quality control is defects occurring during the welding process, which significantly affects inspection cycle times. To address this, the Convolutional Neural Network (CNN) approach with VGG-16 architecture can help classify defects in the welding process. This method not only expedites the defect classification process but also enhances the accuracy of identifying product defects. The stages of implementing this method include dataset preparation, data preprocessing, CNN model design, model training, and performance evaluation. Evaluation results demonstrate that the use of automatic defect detection technology, especially with balanced data scenarios, can significantly improve quality control performance. Accuracy, precision, recall, and F1-score achieve excellent levels, reaching 92%. Thus, this research provides a significant contribution to enhancing production efficiency and improving product quality in the motorcycle manufacturing industry in Indonesia. It is hoped that the use of this technology will assist manufacturing companies in identifying and addressing production defects more effectively, thereby enhancing the overall competitiveness of Indonesia's manufacturing industry.
Comparison of Machine Learning Models for Heart Disease Classification with Web-Based Implementation Ramadhan, Angga Ramda; Saefulloh, Nandang; Utami, Nisa; Diana, Muji; Utomo, Abiyyu Aji Prasetyo; Wicaksana, Yusuf Eka
Journal of Applied Informatics and Computing Vol. 8 No. 2 (2024): December 2024
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v8i2.8744

Abstract

Heart disease has become one of the most concerning diseases in Indonesia according to research published in 2018 by the Health Ministry of Indonesia. Based on said research, 15 out of 1000 Indonesians suffer from heart disease. Furthermore, according to data published by the Health Ministry of Indonesia, 3 million premature deaths (under 60 years old) occurred in 2013 due to heart disease. Therefore, this research aims to develop a web-based system designed to aid health workers in screening for heart diseases and producing early diagnosis. In developing this system, 5 models were evaluated based on performance and the model with the best metrics was selected to be used in the final system. These models were: Logistic Regression, Decision Tree, Random Forest, Naïve Bayes, and K-Nearest Neighbours. SMOTE and ADASYN was also used to deal with imbalanced data, and the resulting balanced data was used as additional training scenarios in order to compare the result with algorithms trained using imbalanced data. Cross validation, accuracy, precision, recall, f1-score, and ROC with AUC were set as evaluation metrics. Results show that Random Forest trained with data balanced using ADASYN achieved the highest AUC score of 0.920. Meanwhile, Logistic Regression scored lowest with an AUC score of 0.500. These results indicate that Random Forest is the most suitable for this system Therefore, Random Forest was selected as the algorithm to be used in the final system. Furthermore, this system has been tested successfully using the black-box method and is ready to be implemented.
Co-Authors Diana, Muji Indra, Jamaludin Juwita, Ayu Ratna Nisa Utami, Nisa Rahmat Rahmat RAMADHAN, ANGGA RAMDA Utomo, Abiyyu Aji Prasetyo Wicaksana, Yusuf Eka