Input analysis
Mouse Control Detection
We collect more than 20 biomarkers such as speed, acceleration, and angular velocity, and use machine learning to analyse the data to distinguish between human and bot behaviour. In case we deal with more sophisticated bots, we employ neural networks to detect subtle anomalies that differentiate human behaviour from bot actions. We employ supervised or semi-supervised methods when we have definitive bot data. For bots without clear labels, we use unsupervised methods.
Keyboard Control Detection
Similar to our mouse control analysis, we gather data from keyboard inputs, including the speed and duration of key presses, along with other relevant metrics. The analysis process for keyboard control mirrors that of mouse control. By reviewing and comparing this data, we differentiate between human and bot behavior.
Through these methods, we efficiently detect cheats across various gaming experiences.
Integration process
Right now we tailor AI integration process for particular game-title based on genre and expected data load. Please contact us for details.