Enhanced Brute Force Attack Detection in Remote Access Security: Integrating ANN and SVM

• Published in: 2024 Asia Pacific Conference on Innovation in Technology (APCIT) pp. 1 - 5
• Main Authors: Kumari, Ankita, Gupta, Deepali, Uppal, Mudita
• Format: Conference Proceeding
• Language: English
• Published: IEEE 26.07.2024
• Subjects: Accuracy; Artificial Neural Network; Artificial neural networks; Brute Force Attack; Brute force attacks; Force; Infrastructure; Passwords; Security; Support Vector Machine; Support vector machines; Technological innovation
• DOI: 10.1109/APCIT62007.2024.10673550

In the area of remote access security, it is very important to find and stop brute force attacks in order to keep important data and resources safe. Criminals often use brute force tactics, which involve trying a lot of different user combinations, passwords, or encryption keys to get into systems or data without permission. This hostile method takes advantage of holes in security measures, which threatens the safety of private data and the system's general stability. Realizing and avoiding strong force attacks is important for keeping digital spaces safe and stopping illegal entry.The proposed methods of Support Vector Machine (SVM) and Artificial Neural Network (ANN). To find likely cases of brute force attacks, the suggested model uses the basic screening features of support vector machines. The model is then made more accurate with Artificial Neural Network improvement. The proposed method gets an impressive 82% success rate by using an iterative process to look at and improve results for a dataset that isn't fair. The proposed model combines the best features of support vector machines and artificial neural networks to find and stop brute force attacks in changing defense situations. When the integrated Support Vector Machines and Artificial Neural Networks and get strong defense against new attack tactics. This keeps important data and systems safe and secure. The results show how important it is to use advanced machine learning techniques to properly reduce hacking risks. Also, the model that has been created shows promise in real-life scenarios, offering a workable answer for companies that want to improve their security against brute force attacks. The SVM-ANN design needs to be constantly monitored and improved in order to keep up with new threats and keep its high level of accuracy. This study provides a significant contribution to the subject of cybersecurity by demonstrating how to actually defend digital assets, prevent brute force assaults, and maintain data integrity.