Optimized Extreme Learning Machine Framework for Efficient Ransomware Detection Using Seagull Optimization-Based Feature Selection

Ransomware is becoming a more serious threat to cybersecurity, and conventional detection techniques frequently fail to maintain up with its changing strategies. To overcome this difficulty, this research proposes an extreme learning machine (ELM) framework enhances the system the efficiency in threat identification. The Static and dynamic feature of files is identified using ELM for analyzing their behavior patterns. The Seagull Optimization Algorithm (SGO) is employed for feature selection refining the dataset by eliminating redundant characteristics, to improve detection process. The system model trained with a Support Vector Machine (SVM) to determine which files are malicious and which are safe. The ELM-SGO-SVM framework is evaluated using performance metrics for identify ransomware threats and it has efficacy in early ransomware detection and mitigation by showing that it greatly outperforms existing techniques. However, this proposed method strengthens the cybersecurity defenses against ransomware threat and ensures better protection for important digital threats.