KeyDetect --Detection of anomalies and user based on Keystroke Dynamics

This falls within the cybersecurity domain, specifically in the areas of user authentication and behavioral biometrics. It represents an innovative approach to enhancing security through the analysis of user behavior.

Authentication focus: The project aims to develop a more secure method of user authentication, which is a core aspect of cybersecurity.
Behavioral biometrics: It utilizes keystroke dynamics, a form of behavioral biometrics, to identify users based on their typing patterns. This is an advanced cybersecurity technique.
Alternative to traditional methods: The project proposes this method as an alternative to conventional two-factor authentication, addressing limitations of current cybersecurity practices.
Anomaly detection: The system is designed to detect anomalies in typing patterns, which could indicate unauthorized access - a key cybersecurity function.
Machine learning application: The project employs various machine learning algorithms such as One-Class SVM, Multiclass SVM, Random Forest Classifier, Artificial Neural Networks (ANN), and Convolutional Neural Networks (CNN) for user classification and anomaly detection, which are increasingly important in modern cybersecurity approaches.
Addressing insider threats: By continuously monitoring typing patterns, this system could help mitigate insider threats, a significant concern in cybersecurity.

Kar, S. (2023) KeyDetect --Detection of anomalies and user based on Keystroke Dynamics. (In review) (Third author)