Page 50 - THEi Student Applied Research Presentations 2025

SMARTTICKETS: AI-DRIVEN SMART
OBJECTIVES
• Develop an efficient online ticketing
system: Build a fully functional
platform that supports event
publishing, user registration, ticket
booking and secure payment.
• Integrate machine learning to
detect scalpers: Use machine
learning models to analyze user
data and automatically identify
potential scalper accounts to
maintain market fairness.
• Implement a fair lottery ticket
purchasing mechanism: Design
and implement a transparent lottery
system for high-demand events to
ensure that all users have equal
opportunities to purchase tickets.
ABOUT THE INVESTIGATOR
Hello. My name is Steve. I’m passionate
about uncovering obscure knowledge
and exploring hardware configurations.
I enjoy watching F1 races, challenging
myself with experiences like bungee
jumping, and finding inspiration in
music. My goal is to innovate in tech,
blending curiosity with professional
excellence. My FYP supervisor is Dr
CHEONG Kai Yuen.
39 Student Applied Research Presentations 2025
TICKET DISTRIBUTION SOLUTION
Poster
LAM Yu
BSc (Hons) in Information and Communications Technology
Department of Digital Innovation and Technology
RESEARCH BACKGROUND
With the rise of online ticketing platforms, scalpers are increasingly hoarding
and reselling tickets, using automated tools or bulk purchasing to profit, e.g the
controversy over the ticket sales of YOASOBI's Hong Kong concert. It went on
sale and sold out quickly.
This not only harms the rights of fans and violates laws. Therefore, utilizing
AI and automatically detect scalper behavior can effectively enhance the
management efficiency of ticketing platforms and reduce the impact of scalpers
on the market.
METHODOLOGY
This project adopts an agile development approach and be carried out in
stages. In the early stages, we design the system architecture and database
model, and use modern web frameworks such as Next.js to develop core
functions, including user management, event publishing, and ticketing systems.
Next, we collect user behavior data to train and evaluate a machine learning
classification model to automatically identify scalper behavior. At the same time,
a fair lottery ticket purchase mechanism is developed and integrated. During
the development process, we continue to optimize database performance and
perform unit and integration tests to ensure the stability, scalability, and security
of the system, and finally deploy it to the cloud platform.
FINDINGS
1. Dual Booking System: Regular events facilitate direct ticket purchases,
while high-demand events utilize a lottery-based allocation method.
2. Anti-Scalping Measures: The system incorporates real-time scalper email
detection using machine learning and dynamically refreshing QR codes to
enhance security.
3. Lottery Management: Automated ticket generation, user email verification,
and friend-based ticket transfers streamline the user experience.
4. Admin Tools: Comprehensive features for event creation, payment
processing, and user monitoring empower administrators to manage the
platform effectively.
5. Anti-scalping measures, including transfer cooldown periods, purchase
limits, and email detection, ensure that genuine fans can participate fairly.
The platform leverages advanced AI technology and an optimized user
experience to maintain market fairness and integrity.
Student Applied Research Presentations 2025