Как создать семантический слой для BI и AI: показатели, измерения и происхождение данных | DataForge

In this video, we demonstrate step-by-step how to create a metrics and dimensions register (MDR) in DataForge—one of the key elements of a modern data architecture. Learn more about the platform at: https://dataforg.ru If a company has multiple BI systems, dozens of reports, and constant debates about how to correctly calculate revenue, profit, customer count, or other metrics, a semantic layer helps bring order. It allows you to create a unified vocabulary of business metrics, capture their definitions, and ensure a consistent understanding of the data by all process participants: analysts, developers, business users, and AI agents. In this video, you'll see the full workflow: • project creation; • database connection; • creating facts; • metric configuration; • creating calculated metrics; • data lineage analysis; • creating dimensions; • creating reference books; • reviewing the completed MDR and its usage principles. This approach is especially useful for organizations implementing BI systems, building enterprise data warehouses, implementing data governance, or beginning to use AI assistants to work with corporate data. DataForge helps transform disparate definitions and Excel files into a single, manageable layer of company data knowledge. Learn more about the platform at: https://dataforg.ru Timecodes: 00:00 Introduction 00:32 Project 00:50 Database Connection 01:04 Creating a Fact 02:00 Creating a Metric 02:53 Creating a Calculated Metric 04:49 Viewing Lineage 05:01 Creating a Dimension 05:47 Creating a Directory 06:40 Review of the Completed RPI 07:01 Conclusion Semantic layer, semantic layer, DataForge, indicator registry, indicator catalog, indicators and dimensions, data lineage, data lineage, data governance, BI, business analytics, DWH, data warehouse, metrics, KPIs, AI agent, text to SQL, corporate data, self-service bi, data catalog