ENHANCING SECURITY USING KEYSTROKE DYNAMICS

The uniqueness and usefulness of keystroke dynamics is an authentication measure using the typing rhythm of the user. A measure to incorporate keystroke dynamics in daily authentication systems has not been made. User’s identity is analyzed using keystroke through their way of typing on a computer keyboard. Keystroke is a genuinely software centric solution and can be used for long samples of text which further can be used to increase the robustness of the method. Keystroke dynamics is an underappreciated system and can be used extensively as a robust authentication tool either individually or as a hybrid system. Knowing the cost effectiveness and efficiency we discovered that a much better use of keystroke analysis for authentication needs to be used. We also introduce keystroke dynamics based authentication and present distance metric or scoring based machine learning models. The proposed work develops a Key logger to record the keystrokes of the user and stores it. A custom built feature extractor for extracting features from key logger data is developed and also incorporates different classification models to classify and measure the similarity. Finally authentication of the user is done based on the results. Keywords— keystroke dynamics; authentication; security;