Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Journal of Innovation Information Technology and Application (JINITA) Informatics and Software Engineering (ISE)
Daehyon Kim
Chonnam National University
Computer Science & IT Decision Sciences, Operations Research & Management Engineering

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

Found 2 Documents
Search

 1 
Traffic Image Analysis Based on Stacked Denoising Autoencoder Neural Network Daehyon Kim
Journal of Innovation Information Technology and Application (JINITA) Vol 5 No 2 (2023): JINITA, December 2023
Publisher : Politeknik Negeri Cilacap

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35970/jinita.v5i2.2133

Abstract

This study aims to explore major neural network models - Stacked Denoising Autoencoder (SDAE), Deep Belief Network (DBN), Backpropagation - that have recently garnered attention and propose the most suitable and reliable artificial neural network model for real-time road traffic information collection. In this study, to enhance the reliability of experimental results, numerous experiments were conducted under identical conditions (such as parameter values and network configuration) by setting different initial values for the weight vector. The results of the experiments were statistically validated to draw conclusions. The research results showed that the SDAE model exhibited the most superior performance, while the accuracy of the DBN was somewhat lower compared to the SDAE model. On the other hand, the Backpropagation model demonstrated a relatively low predictive accuracy compared to both models, particularly showing a significant influence of the initial values
Unified Predictive Modeling: Enhancing Accuracy with DBNs, Fuzzy ARTMAP, and SVMs Daehyon Kim
Informatics and Software Engineering Vol. 3 No. 1 (2025): June 2025
Publisher : SAN Scientific

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58777/ise.v3i1.425

Abstract

Amid the global surge in artificial intelligence, the field of machine learning is advancing rapidly, and selecting the most suitable algorithm for prediction tasks remains a crucial challenge. This paper introduces a novel ensemble model that combines three machine learning algorithms—Deep Belief Networks (DBNs), Fuzzy ARTMAP, and Support Vector Machines (SVMs)—to enhance predictive performance. Each machine learning model possesses unique strengths, and by integrating these models, it is possible to overcome individual model limitations and achieve more accurate and reliable predictions. DBNs excel at learning hierarchical representations and capturing complex patterns, Fuzzy ARTMAP is proficient in handling imprecise and ambiguous data, and SVMs are renowned for their robustness in high-dimensional spaces. Thus, the integrated framework leverages the complementary strengths of each model while mitigating their weaknesses. In this study, the proposed ensemble model's predictive power was validated through experiments on image data collected from actual construction sites for construction automation research. The prediction performance of the proposed ensemble model was evaluated and compared with that of individual models such as DBNs, Fuzzy ARTMAP, and SVMs, demonstrating its superiority. The experimental results showed that the proposed model outperformed each individual algorithm in terms of prediction accuracy, clearly illustrating the effectiveness of the ensemble approach.
Co-Authors