Investigating Artificial Intelligence-based Solution for Protecting Online Gaming Servers Against Anomalies

The proliferation of affordable mobile phones and computers has led to an increasing prevalence of individuals engaging in game-playing activities. This trend has been particularly pronounced in the realm of online gaming, with a notable surge seen over the last decade. The consumers have a preference for engaging in sophisticated gaming experiences inside virtual environments. Numerous firms are offering their services in this particular domain. With the escalating competitiveness, gaming service firms are susceptible to being attacked by hackers for various malicious activities such as ransomware attacks and data theft. It is important to implement measures to safeguard the gaming web server from potential assaults in order to provide customers with a secure environment for uninterrupted access to gaming services. This study presents a machine learning-driven approach to identify the genuineness of communication at an early stage. The proposed model utilizes a pre-trained algorithm to assess whether incoming data packets are malicious or benign. The proposed framework is evaluated by applying it to anomalies in a web server dataset, using classification machine learning methods. The obtained results are then compared based on several metrics like accuracy, F1 score, ROC Curve, and precision. The use of sophisticated machine learning models such as Random Forest, Decision Tree, and Gaussian NaiveBayes has yielded valuable insights pertaining to the detection of abnormalities, atypical patterns, and deviations from expected behavior inside gaming server ecosystems.