• Dr.P.Thamarai
• Kuridi Dwaraka Sai Ganesh
• Kosana Manohar
• Kommisetti Sai Surya Manikanta
• Kuracha Sri Satya Murali Raghava

This project develops a Zero-Input Behavioral Biometric Authentication system that identifies users based on their natural mouse drift and scroll patterns. Unlike passwords or tokens, this method works silently in the background and does not need the user to do anything extra. It provides continuous authentication, which means the system can keep checking if the correct user is active even after login. The system records different mouse activities such as speed, acceleration, drift direction, curve in movement, pause times, scroll frequency, and scroll speed. From these, special behavioral features are extracted to represent the unique way a person interacts with the computer. To make the data reliable, preprocessing steps like filtering, scaling, and normalization are applied. For building the model, a One-Class Support Vector Machine (SVM) is used, which is effective because it requires training only with the genuine user’s data and does not need attacker data. The SVM creates a profile of the user’s behavior, and during real use, the system compares live mouse actions with the stored profile. The main advantage of this approach is that it adds an extra layer of unobtrusive security on top of normal login systems. Even if an attacker steals the password, they can be detected if their mouse behavior does not match the real user. The method is light, fast, and can be scaled to personal computers, corporate networks, and secure environments. This project also reduces dependency on remembering passwords, lowers chances of insider misuse, and provides a real-time alert mechanism for abnormal behavior. It is also resistant to spoofing attacks because human mouse movement is difficult to mimic exactly. The One-Class SVM ensures adaptability and defense against spoofing, while real-time detection makes the system practical. This creates a step toward next- generation authentication, where the system continuously checks identity using natural behavior, instead of depending only on passwords.

"CONTINUOUS BEHAVIORAL AUTHENTICATION THROUGH CURSOR DYNAMICS AND SCROLLING BEHAVIOR", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 11, page no.d319-d326, November-2025, Available :http://www.jetir.org/papers/JETIR2511335.pdf

"CONTINUOUS BEHAVIORAL AUTHENTICATION THROUGH CURSOR DYNAMICS AND SCROLLING BEHAVIOR", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 11, page no. ppd319-d326, November-2025, Available at : http://www.jetir.org/papers/JETIR2511335.pdf