Personal portfolio website
This is my personal portfolio website developed using Django, HTML, CSS, and JavaScript. It comprehensively showcases my professional journey, featuring detailed sections on my work experience, educational background, and a curated collection of past projects.
Ai-Traffic Control System
Ai-Traffic Control System is my dissertation project when i was in the university, aiming to reduce traffic conjection in urban area. This project leverages YOLO v8 for real-time traffic object detection, integrated with a GRU (Gated Recurrent Unit) model to predict and optimize traffic flow. Utilizing SUMO as the simulation environment, the research demonstrated a significant 16% reduction in urban traffic congestion
QR Code Detection System
QR Code Detection System for Construction Site Safety Created a Python-based QR code detection system to monitor equipment safety, collaborating with China Mobile HK. Implemented image transformation and QR code decoding techniques to track over 100+ employer equipment in real-time, enhancing workplace safety and inventory management.
Tools Detection System for Construction Site Safety
Tools Tracking System for Correctional Services Developed an advanced object detection system for Hong Kong Correctional Services using Python, YOLO, and VGG16. The system accurately verifies and tracks tools borrowed by 10,000+ prisoners, significantly improving inventory control and institutional safety.
Disease Treatment Response Prediction
Developed a machine learning model using Python to predict treatment responses for diseases. Utilized pandas and sklearn for feature selection, and applied k-Fold cross-validation to enhance model performance and accuracy.
Restaurant Booking Chatbot
Created an advanced chatbot using natural language processing techniques. Implemented sentiment analysis and TF-IDF vectorization with cosine similarity to improve dialogue detection and response accuracy, demonstrating sophisticated NLP capabilities.
Water Pump Functionality Prediction
Conducted comprehensive predictive modeling for water pump functionality using Python. Performed Exploratory Data Analysis (EDA), data cleaning, and preprocessing, including correlation matrix analysis. Utilized K-Fold cross-validation to assess and improve model performance.