Development of Intelligent Frameworks for Brute Force Intrusion Detection using Large Language Models
Abstract
As people utilize more digital devices and networks, cyber dangers like brute force attacks that take advantage of poor authentication techniques are becoming more common. Signature-based or classical machine learning algorithms are often not very good at uncovering new and unexpected attack patterns with low false positives. This paper suggests creating a smart hybrid system that uses ML and LLMs to find brute-force attacks. The method starts with speedy initial detection through data preprocessing, feature engineering, and categorization based on machine learning. Then it moves on to semantic analysis led by LLMs, which makes it even easier to understand the context and make decisions. Experimental results reveal that the suggested hybrid model works better than solo models since it lowers false alarms and raises accuracy, precision, recall, and F1-score. Adding LLMs can help you comprehend and evaluate complicated attack behaviors more easily.
How to Cite
Komal, Dr. Shailesh Kumar , "Development of Intelligent Frameworks for Brute Force Intrusion Detection using Large Language Models", Vol. 4, Issue 4, 24-07-2026, pp. 50-59. DOI: 10.5281/zenodo.22041266
