Abstract

Enterprise data warehousing (EDW) systems consolidate structured and unstructured organizational data from multiple sources to support strategic decision-making. However, the sheer volume, variety, and velocity of enterprise data create challenges in extracting actionable insights efficiently. Knowledge discovery (KD) models provide systematic approaches for identifying patterns, relationships, and trends within large-scale datasets, enabling improved decision analytics and business intelligence. This paper explores knowledge discovery frameworks applied to enterprise data warehousing, focusing on data preprocessing, feature extraction, pattern mining, and predictive analytics. Techniques include statistical analysis, machine learning algorithms, clustering, association rule mining, and classification models, which facilitate informed decision-making and resource optimization. The study evaluates KD models based on their ability to handle large datasets, detect meaningful patterns, and support predictive and prescriptive analytics. Experimental analysis demonstrates that integrating knowledge discovery techniques into enterprise data warehousing significantly improves decision support, enhances operational efficiency, and reduces response time for analytics queries. Additionally, hybrid approaches combining statistical and AI-driven models are highlighted for their effectiveness in complex decision environments. This research emphasizes the importance of adaptive, scalable, and interpretable knowledge discovery models for enhancing enterprise analytics capabilities and driving data-informed organizational strategies.