Spam messages create many problems in digital communication systems. They cause inconvenience, security issues, and a waste of time. Rule-based spam filters are easy to understand, but they cannot change according to new spam patterns. Machine learning methods give better accuracy, but they are difficult to understand because they work like black boxes. This paper explains a hybrid intelligent system for spam detection. The system combines rule-based filtering and machine learning techniques. The rule-based part identifies clear spam messages using predefined rules, while the machine learning model learns patterns from labeled data. The proposed system is tested using a standard public spam dataset. The results show that the hybrid system gives better accuracy than machine learning models alone and also provides a clear explanation for decisions. Therefore, the hybrid approach is suitable for educational and real-world applications.

Keywords: Spam Detection, Hybrid Intelligent Systems, Rule-Based Systems, Machine Learning, Explainable AI