The massive increment of the free-to-play (F2P) mobile gaming market has seen a high rise in various monetization strategies, many of which takes advantage of player behavior and psychology. Some of these strategies, such as loot boxes, pay-to-win mechanics, and artificial scarcity, have come under scrutiny for being deceptive. This research develops a machine learning model to detect such deceptive monetization patterns by analyzing features related to player behavior, game mechanics, and in-game transactions. Using a synthetic dataset of player profiles, the model was trained to classify games as deceptive or non-deceptive. The performance of multiple machine learning algorithms was evaluated using metrics like accuracy, precision, recall, and F1-score, with the Random Forest model yielding the best results. The findings offer a robust framework for detecting deceptive patterns, contributing to fairness and transparency in the gaming industry while also providing insights for future regulatory efforts.

Authors: Oduola Tunde Yusuf, O. Owolafe, Olaniyi A. Ayeni

Published in: International Conference for Internet Technology and Secured Transactions (ICITST-2024)

• Date of Conference: 4-6 November 2024
• DOI: 10.20533/ICITST.2024.0011
• ISBN: 978-1-913572-76-1
• Conference Location: St Anne’s College, Oxford University, UK