As the days go by, the need for security keeps increasing; whether for humans, data, climate and almost everything in existence. However, despite the increase in the need to protect data, information and the corresponding devices, the Personal Identification Number (PIN) is still the most widely used approach on mobile devices. This increased requirement for protection is evidenced by various issues that continue to arise based on intrusion, theft and unlawful access to classified information. It is important to note that apart from secret-knowledge (that is PIN), biometrics and tokens are useful for security even though token seldom authenticates because it is expected to be with the authentic owner. Biometrics on the other hand makes use of the individual himself. This research sought to address the issue of securing data on mobile devices by employing some Machine Learning algorithms. The K-Nearest Neighbours (KNN), Decision Trees (DT) and Multinomial Logistic Regression (MLR) classification algorithms were used to train and test the typing patterns of several individuals who volunteered to type a static passphrase. After carrying out the different experiments, the predictions given by Decision Trees were the most accurate of all the three base classifiers used with an accuracy value of 99.92%.

Keystroke Dynamics Typing Pattern Confusion Matrix