Anomaly detection plays a crucial role in various domains where identifying rare yet significant deviations is essential. While Isolation Forest (ISF) is a widely used unsupervised method, it often suffers from false alarms and limited interpretability. To address these challenges, we propose an enhanced framework that integrates a knowledge base derived from Association Rule Mining (ARM). This integration improves anomaly detection by leveraging automatically generated rules while allowing business defined rules to be incorporated, making the model more adaptive and context aware. Our approach consistently improves both AUC and precision, with average gains ranging from 0.13% to 0.37% in AUC and 0.5% to 2.51% in precision across datasets and configurations of our framework. A key advantage of our framework is its ability to reduce false positives, resulting in a more reliable and precise anomaly detection system. Additionally, it enhances interpretability, allowing users to easily identify rule violations at the record level. Beyond improving ISF’s detection capabilities, this framework offers a scalable and flexible solution that can adapt to diverse business needs.