Knowledge Management Group

Sidney Tsang is currently completing his PhD at the University of Auckland, which began on April 2011. He previously completed a Bachelor of Engineering degree, specialising in Software Engineering, also at the University of Auckland.

The work in my PhD has to do with fraud detection in online aution sites such as eBay or TradeMe. I have developed an agent-based simulation of online auctions, modelled on data collected from TradeMe, which generates realistic synthetic auction data. The goal is to be able to develop and rigorously evaluate detection methods for different types of online auction fraud.

I am currently working on extending the simulator to produce synthetic data containing different types of fradulent behaviour. In addition, the simulation will be extended to model additional network-level features, such as how users choose which other users to interact with. This synthetic data will be used as a dataset for supervised learning techniques such as neural networks or decision trees to develop improved fraud detection methods.

Sidney Tsang, Yun Sing Koh, Gillian Dobbie, Shafiq Alam: Detecting online auction shilling frauds using supervised learning. Expert Syst. Appl. 41(6): 3027-3040 (2014)

Sidney Tsang, Yun Sing Koh, Gillian Dobbie: Finding Interesting Rare Association Rules Using Rare Pattern Tree. T. Large-Scale Data- and Knowledge-Centered Systems 8: 157-173 (2013)

Sidney Tsang, Gillian Dobbie, Yun Sing Koh: Generating Realistic Online Auction Data. Australasian Conference on Artificial Intelligence 2012: 120-131

Sidney Tsang, Gillian Dobbie, Yun Sing Koh: Evaluating Fraud Detection Algorithms Using an Auction Data Generator. ICDM Workshops 2012: 332-339

Sidney Tsang, Yun Sing Koh, Gillian Dobbie, "RP-Tree: Rare Pattern Tree Mining", DaWaK 2011: 277-288