Published · 2014
Identifying high-risk online gamblers: a comparison of data mining procedures
International Gambling Studies, 14(1), 53-63
Plain-language summary
Compares data-mining methods for identifying high-risk online gamblers from behavioral records. Strong performance on training data did not consistently carry over to new observations, and even the best-performing method missed many high-risk players. The study emphasizes validation beyond the data used to build a model.
Cite this work
Philander, K. (2014). Identifying high-risk online gamblers: a comparison of data mining procedures. International Gambling Studies, 14(1), 53-63.