The Server-Side Request Forgery (SSRF) exploit enables an attacker to send responses from a server that is weak to other internal and external systems. This is a potent target for attackers, which can be employed to obtain private information, run malicious scripts, or even initiate additional assaults. SSRF operates by using a server's vulnerability to make requests on the attacker's account. The most effective strategy to guard against server vulnerabilities is to routinely check for unusual behaviour and maintain the system updated with the most recent security updates. In this approach, server-side request forgery (SSRF) vulnerabilities are determined by determining the probability of incidents associated with the attack exposure using machine learning. Suspicious activity within the server logs is detected and monitored to attain server-side forgery attacks. Unusual requests, failed login multiple attempts, and also other suspicious activity indicate the attack is in progress. In this research, the dataset is taken from the CISA standard of known exploited vulnerabilities with different attributes. Ensure that the server-side code is properly secured and that server logs are monitored and responded to appropriately for any incidents based on the likelihood of an attack succeeding, which can be significantly reduced.