Gurninder Singh

Computer Networking

Completed: [June, 2018]

• Covered LAN/WAN concepts, IP addressing, subnetting, routing & switching basics
• Hands-on experience using Cisco Packet Tracer for router/switch configuration
• Learned networking protocols (Ethernet, TCP/IP, ARP, DHCP, DNS) and OSI model fundamentals

Certified in Cybersecurity (CC) — (ISC)²

Completed: [July, 2023]

• Entry-level cybersecurity certification covering security principles, network security, access controls, risk management, and incident response
• Validates foundational knowledge in cybersecurity as aligned with (ISC)² industry standards

Medicare – Health Services Website
Technologies: PHP, MySQL, HTML, CSS, JavaScript

• Designed and developed a dynamic web application for managing patient information and healthcare services
• Implemented user registration, login system, appointment scheduling, and database-driven content
• Built secure backend with MySQL to store patient records, doctor availability, and appointment history
• Ensured responsive design and basic input validation for enhanced user experience and data integrity

GST Calculator – Tax Computation Tool
Technologies: Java (Core), Java Swing 

• Developed a standalone application in Java to calculate Goods and Services Tax (GST) based on Indian tax slabs
• Allowed users to input item price and GST rate to compute CGST, SGST, and total payable amount
• Implemented error handling and input validation to ensure accurate calculations

Mitigating Credit Card Fraud In European Countries

Tools & Techniques: Python, Scikit-learn, HMM, Logistic Regression, ANN, Decision Trees

• Conducted a comparative analysis of machine learning models (HMM, Logistic Regression, ANN, Decision Trees) for credit card fraud detection using a large transaction dataset.
• Achieved 99.97% accuracy with ANN and Decision Tree models; Logistic Regression reached 99.83%, while HMM showed 89% accuracy with a 0.93 precision and 0.83 recall, indicating strength in identifying sequential patterns.
• Analyzed model behavior on imbalanced datasets and proposed improvements via ensemble methods, deep learning (LSTM, RNN), and real-time data testing.
• Focused on model scalability, computational efficiency, and enhancing fraud detection adaptability across diverse datasets.