Behavior Based Approach for User Validation Using Machine Learning

Abstract

In today’s world, security and privacy are the most important considerations for internet users in every application domain. The Internet and applications are currently involved in nearly every area of our day-to-day lives. Users can access lots of things just by using internet connectivity. So, user authentication and validation are very important in this digital world. For a long time, to authenticate and validate the user, traditional systems such as pins, passwords, tokens, and two-factor authentication were used. The majority of applications employ One-Time- Password (OTP) as a two-factor authentication method. OTP shared over SMS, email and third-party applications is valid for a limited time. Also, sometimes in order to receive the OTP, users need a device that has network connectivity. Users may not get or be unable to obtain an OTP for a range of factors, including network issues, severe network traffic, and inability to use a phone.

The focus of this paper is to find the optimum method for constructing a validation system that employs ml algorithms to authenticate individuals relying on their keyboard behaviour. In this research we applying different similarity algorithm and find the suit- able approach. Moreover, we discussed about the data set, technique and results with achieving mentioned goal.