AI-Enhanced Anomaly Detection and Automated Root-Cause Analysis for Enterprise Data Platforms

Enterprise data platforms are increasingly complex, making timely anomaly detection and root-cause diagnosis difficult using conventional rule-based monitoring, which typically detects only <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{6 0 - 7 0 \%}$</tex> of critical incidents and generates excessive false alerts. To address these limitations, this study proposes an AI-enhanced framework for real-time anomaly detection and automated root-cause analysis in enterprise data platforms. The proposed approach employs an ensemble learning architecture that integrates statistical techniques, isolation forests, long short-term memory (LSTM) networks, and transformerbased models to capture both temporal and contextual system behaviors. Automated root-cause analysis is enabled through causal inference graphs and contextual embeddings, providing interpretable diagnostic insights. The framework was evaluated over 18 months across three enterprise environments processing more than 10 TB of telemetry data daily. Experimental results demonstrate 94.3 % anomaly detection accuracy with a 2.1 % false positive rate, while root-cause identification achieved 89.7 % accuracy. The system reduced mean time to detection by 73 % and mean time to resolution by 68 %, delivering scalable, explainable, and cost-effective monitoring with annual savings of approximately USD <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{2. 3}$</tex> million per deployment.