Predicting Cheaters in PlayerUnknown’s Battlegrounds (PUBG) using Random Forest Algorithm

Nurin Alya, Haris (2023) Predicting Cheaters in PlayerUnknown’s Battlegrounds (PUBG) using Random Forest Algorithm. [Final Year Project Report] (Unpublished)

	PDF Nurin Alya Binti Haris (24pgs).pdf Download (568kB)
	PDF (Please get the password by email to repository@unimas.my , or call ext: 3914 / 3942 / 3933) Nurin Alya Binti Haris (fulltext).pdf Restricted to Registered users only Download (2MB)

Abstract

PlayerUnknown's Battlegrounds (PUBG) has become a massively popular online video game, attracting a significant number of players of all ages. However, the prevalence of cheating presents an important challenge to maintain a fair gaming environment. This study aims to address this issue by developing a reliable prediction model to identify potential cheaters in PUBG matches. The research methodology involves the collection of a comprehensive data set from Kaggle containing a variety of gameplay features. Patterns and relationships between input variables and cheating behaviours are analysed through the application of supervised learning techniques, specifically a classification model. The primary goal is to use the Random Forest algorithm, an effective machine learning technique, to predict instances of cheating based on the behavioural patterns of participants. Utilising the selected features, the Random Forest model is then trained to produce a robust prediction model. The evaluation of the prediction model demonstrates its accuracy, precision, recall, and F1 score, demonstrating its capacity to identify potential cheaters in PUBG encounters. The developed prediction model obtains an impressive accuracy of 95.7%, demonstrating its reliability in distinguishing cheaters from non-cheaters. The outcomes of this research contribute to the advancement of cheat detection mechanisms in online gaming. The developed prediction model can be integrated into existing systems to improve cheat detection capabilities and promote an enjoyable and fair gaming experience for all PUBG participants

Actions (For repository members only: login required)

View Item