A study of risk-based authentication system in cyber security using machine learning

The optimum authentication method is determined by the user's risk profile, which is created using context- and behavior-based data from the user's device, finger print, one-time password, and other characteristics. Hacking and security breaches of online accounts, including social networking and web ac- counts, are very common in today's society. We suggest a Risk Based Authentication System utilizing Machine Learning to stop this. For the protection of data and money in this internet environment, security is a worry. Numerous parameters are researched and taken into consideration in the paper in order to solve the issue. These variables determine whether to grant the user permission or not. The gradients descent method is used to verify the user. Previous literature is re- viewed with technical details of the system before conclusion.