News

A recent article describes a novel approach using Machine Learning to detect large-scale fraudulent online sellers by recognising give-away signs in how they conduct their digital marketing. The strategy is scalable. The authors employ a supervised learning approach to classify postings as fraudulent or real based on past data from buyer and seller behaviours and transactions on a popular online marketplace platform. They combine bespoke data preprocessing procedures, feature selection methods, and state-of-the-art class asymmetry resolution techniques to search for aligned classification algorithms capable of discriminating between fraudulent and legitimate listings. Their best detection model obtained a recall score of 0.97 on the holdout set and 0.94 on the out-of-sample testing data set, based on 45 selected features.