Crime Pattern Detection Utilizing Power BI Visualizations on the Microsoft Fabric Data Platform With the Public data.police.uk Dataset

This paper presents a big data-driven platform for crime pattern detection using Power BI visualizations on Microsoft Fabric, built upon the public data.police.uk dataset. Law enforcement agencies require scalable analytics to extract actionable insights from large, complex datasets. We implemented a Data Lakehouse architecture to process around 8,500 crime data files i n C SV format from multiple regions, with Python-based metadata cataloging for structured access to crime outcomes, stop-and-search records, and street-level incidents. Dataflows and Notebooks in Fabric addressed regional inconsistencies and enabled efficient data transformation. Power BI reports provided intuitive and interactive visualizations for exploring geographic and temporal crime trends. Performance testing demonstrated up to 40% faster query response times compared to traditional warehouses, and regional crime analysis that previously took days was completed within hours. The results indicated that the platform scaled efficiently while maintaining stable performance under growing data volumes. Our approach demonstrates how unified analytics and visualization environments can democratize access to crime data insights, supporting evidence-based policing and public safety decision-making.