With the increasing sophistication of web spoofing attacks, there is a growing need for robust mechanisms to secure clients against URL-based phishing attacks. This research proposes a novel approach utilizing Machine Learning, specifically Multi-Layer Perceptron (MLP), to enhance the detection capabilities against web spoofing attacks. The proposed system leverages a dataset comprising legitimate and malicious URLs, utilizing features derived from URL structures and content. The MLP model is trained on this dataset to learn patterns and characteristics indicative of phishing attempts. The trained model is then employed to classify URLs in real-time, effectively identifying and preventing potential web spoofing attacks. The system explores the effectiveness of the MLP model in comparison to traditional methods, demonstrating superior accuracy and efficiency in detecting URL-based phishing attempts. Additionally, the system adapts to evolving attack techniques by continuously updating its knowledge base, ensuring a proactive defense against emerging threats. The results indicate that the integration of Machine Learning, particularly MLP, provides a reliable and scalable solution for securing clients against web spoofing attacks. This approach holds promise for enhancing cybersecurity measures, safeguarding users from the ever-evolving landscape of phishing threats in the digital realm.