ITEJ (Information Technology Engineering Journals)

Optimizing Loan Approval Processes with Support Vector Machines (SVM)

2025-02-11T12:59:13SE Asia Standard Time

Loan approval is a critical process in banking, requiring accurate assessment of borrower risk to minimize defaults while maintaining customer satisfaction. This study explores the optimization of loan approval processes using Support Vector Machines (SVM), a robust machine learning method known for its effectiveness in classification tasks. We utilized a dataset comprising historical loan applications, incorporating features such as credit score, income level, debt-to-income ratio, and employment history. The SVM model was trained and evaluated using cross-validation techniques to ensure generalizability. Our results demonstrate that SVM outperforms traditional statistical methods in predicting loan approval decisions, achieving higher accuracy and a significant reduction in false positives. Furthermore, feature importance analysis revealed that credit score and debt-to-income ratio are the most influential factors in the model's decision-making process. By integrating the optimized SVM model into existing banking workflows, institutions can streamline their approval processes, reduce operational costs, and improve customer experience. This study highlights the potential of SVM in modernizing decision-making frameworks in the banking sector, paving the way for further adoption of advanced machine learning techniques in financial services.

Design and Implementation of Network Security Systems on Virtualized Networks

2025-02-11T12:59:13SE Asia Standard Time

This report, entitled "Design and Implementation of Network Security Systems on Virtualized Networks," was prepared to fulfill the final assignment of the Network Security course at the Bacharuddin Jusuf Habibie Institute of Technology (ITH). This research aims to design, implement and identify network security vulnerabilities in a virtualization environment using Proxmox Virtual Environment (Proxmox VE) in VirtualBox. The research results show that Proxmox VE in VirtualBox is less successful in optimizing software-hardware resources by implementing security mechanisms such as firewalls, encryption, IDS/IPS, VPN, and IAM. Even though it has several shortcomings, Proxmox VE has proven to be effective in managing virtual networks safely and efficiently when carried out outside of VirtualBox. This research also provides practical experience for students in implementing and identifying network security vulnerabilities, preparing them for real-world challenges.

Cross-Domain Transfer Learning: Enhancing Deep Neural Networks for Low-Resource Environments

2025-02-11T12:59:13SE Asia Standard Time

Deep neural networks (DNNs) have achieved remarkable success in various domains; however, their performance often relies heavily on large-scale, high-quality labeled datasets, which are scarce in low-resource environments. Cross-domain transfer learning has emerged as a promising technique for adapting pre-trained models from data-rich source domains to low-resource target domains to address this limitation. This study explores innovative strategies to enhance the performance and applicability of DNNs through cross-domain transfer learning, focusing on challenges such as domain disparity, data scarcity, and computational constraints. We evaluate several transfer learning approaches, including feature-based transfer, parameter fine-tuning, and adversarial domain adaptation, across diverse healthcare, agriculture, and natural language processing applications. Experimental results demonstrate significant improvements in model accuracy and generalization in low-resource environments, with accuracy gains of up to 20% compared to models trained from scratch. Additionally, we analyze the impact of transfer learning on reducing training time and computational requirements, making it a viable solution for resource-constrained settings. Despite its potential, the study highlights critical challenges, including negative transfer, model interpretability, and ethical considerations in domain transfer. Addressing these issues, we propose a framework for selecting optimal source domains and enhancing model robustness through hybrid techniques and unsupervised learning. This research emphasizes the transformative potential of cross-domain transfer learning in bridging the gap between data-rich and low-resource environments, paving the way for more equitable and efficient applications of deep learning technologies worldwide.

Integrating IoT and Artificial Intelligence for Sustainable Smart City Development: A Case Study Approach

2025-02-11T12:59:13SE Asia Standard Time

Rapid urbanization growth has intensified the demand for sustainable smart city solutions that optimize resource management, improve citizens’ quality of life, and reduce environmental impact. Integrating Internet of Things (IoT) technology with artificial intelligence (AI) offers transformative opportunities to address these challenges by enabling real-time data collection, analysis, and decision-making. This study explores the potential of integrating IoT with AI for sustainable smart city development, using a case study approach to examine its application in diverse urban domains, including energy management, transportation, waste management, and public safety. The research highlights innovative IoT-enabled systems such as smart grids, intelligent traffic control, predictive waste collection, and AI-driven surveillance, demonstrating their ability to improve efficiency and sustainability. Case studies from globally recognized smart cities such as Singapore, Barcelona, and Copenhagen illustrate the benefits and challenges of adopting these technologies. Key findings reveal significant improvements in energy efficiency (up to 25%), reduced traffic congestion (up to 30%) and optimised waste management (up to 40%). However, challenges such as data privacy, interoperability and high implementation costs remain barriers to large-scale deployment. This study proposes a framework to address these issues, emphasizing collaborative governance, robust cybersecurity measures and scalable infrastructure design. The findings underline the transformative potential of integrating IoT and AI to achieve sustainable urban development, offering practical insights for policy makers, urban planners and technology developers. This research contributes to driving smart city initiatives by bridging technological innovation with sustainability goals, paving the way for more resilient and liveable urban environments.

A Review: Development of an IoT-Based Smart Home Monitoring System for the Comfort of People with Disabilities

2025-02-11T12:59:14SE Asia Standard Time

The advancement of the Internet of Things (IoT) has revolutionized smart home technology, providing enhanced comfort, security, and accessibility, particularly for people with disabilities. This review explores the development of IoT-based smart home monitoring systems designed to improve the quality of life for individuals with mobility, sensory, or cognitive impairments. By integrating IoT sensors, automation, and artificial intelligence (AI), smart home systems can provide real-time monitoring, adaptive control of household appliances, and emergency response mechanisms. The study highlights key technologies such as voice-controlled assistants, smart sensors, and remote accessibility features, which enable seamless interaction with home environments. Additionally, challenges related to data privacy, security risks, and affordability are discussed, along with potential solutions. The findings suggest that IoT-enabled smart home systems significantly enhance the independence and well-being of individuals with disabilities, emphasizing the need for continued innovation and policy support in this field. Keywords— Disability, Smart Home, Google Assistant, IoT.