Decision intelligence platforms are typically based on 3 main pillars: Data fusion from multiple sources: A decision intelligence solution must promote the collaboration of both tactical and high-level decision-makers, often across teams, so that all stakeholders have access to relevant databases and sources in a single container to facilitate cross-referencing and validation of data. A decision-making platform can thus make data-backed decisions more accurate and efficient by optimizing caseloads resources, and ultimately improving ROI. Advanced analytics: Once data is fused, a decision intelligence solution can apply advanced analytics and machine learning to highlight patterns and anomalies, find hidden connections, and assess risks. These advanced analytics enable analysts to reach insights that would otherwise remain unattainable, including automated risk scoring based on suspicious indicators. Alerts based on risk assessment can prompt analysts to dig deeper into entities or events that were previously undetected and to gain insights that would otherwise be missed. Flexible data modeling: Every organization is unique and must model data according to its own workflows and procedures. Some decision intelligence solutions are fitted with out-of-the-box data models which may be limiting for certain organizations with complex needs. Flexible data modeling is necessary to ingest and fuse new data sources as well as tailor existing machine learning models to suit the needs of a specific task. Having a dynamic data modeling capacity allows users to adjust machine learning models based on specific insights and requirements. 1 3 4 Decision intelligence platform 1 2