Paper Title
Machine Learning Based Network Anomaly Detection with Explainable AI

Abstract
This paper presents a machine learning–based approach to anomaly detection in network traffic using the Random Forest algorithm for efficient classification. Unlike traditional intrusion detection systems that rely on rule- based techniques, the proposed framework improves detection capability by learning patterns from network data and accurately classifying normal and anomalous traffic. The proposed system was evaluated using synthetic network traffic as well as the CICIDS2017 benchmark dataset. Experimental results demonstrate that the model achieves strong performance in terms of accuracy, precision, recall, and F1-score. Comparisons with baseline models, including One-Class SVM and other machine learning approaches, show that the Random Forest–based method provides a Reliable balance between detection performance and computational efficiency. The system is implemented with a lightweight Tkinter- based GUI, making it accessible for both researchers and practitioners. Designed with constrained environments such asWirelessSensor Networks(WSNs) inmind,this solution emphasizes low computational cost and efficient performance. The study contributes by demonstrating the effectiveness of machine learning techniques for practical and deployable network anomaly detection in cybersecurity applications. Keywords - Anomaly Detection, Isolation Forest, SHAP, Explainable AI, Network Security, Wireless Sensor Networks, Machine Learning, Data Visualization.