Integration of Geo-Spatial Business Intelligence and GSTIN Data for Identifying Fake Registrations and Non-Registered Commercial Entities

The integrity and efficiency of the Goods and Services Tax (GST) administration in India depend upon accurate taxpayer registration, geographical verification of business premises, and systematic compliance monitoring. Despite digital registration systems under GSTN, challenges such as fake registrations, shell entities, non-filers, and unregistered commercial establishments continue to cause revenue leakage and administrative inefficiencies. This doctoral research proposes a geo-spatial business intelligence framework that integrates pincode-wise commercial location data with official GSTIN registration and return filing databases to enhance risk-based inspection and enforcement.

The study employs a Python-based data extraction model using map-based APIs to collect structured commercial metadata, including business name, trade category (shops, restaurants, textiles, jewellery, wholesale establishments, etc.), address components, latitude, and longitude coordinates. Data is extracted within defined geographic radii around specific pincodes to ensure jurisdictional clarity and facilitate administrative distribution of inspection responsibilities. The extracted geo-referenced business records are standardized and systematically compared with GSTIN registration data at the pincode level.

The analytical framework identifies three principal risk categories: (i) GST-registered entities mapped to suspicious or identical geo-coordinates suggesting potential shell or fake registrations; (ii) registered businesses exhibiting persistent non-filing or irregular return submission patterns; and (iii) commercially active establishments visible in geo-spatial datasets but absent from GSTIN records, indicating potential non-registration. Spatial density mapping and coordinate clustering are applied to detect abnormal concentrations of similar trade activities within limited geographic boundaries.

A structured risk-index mechanism is developed combining geographic validation, registration metadata consistency, filing behavior indicators, and trade-category mapping. The pincode-wise analytical output enables structured case allocation to inspection-wing officers, providing measurable, location-justified evidence for field verification. This approach improves transparency in enforcement prioritization and reduces discretionary audit selection.

The findings demonstrate that integrating geo-spatial commercial intelligence with GST administrative records significantly enhances detection of fake registrations and unregistered entities compared to conventional compliance screening methods. The proposed framework offers a scalable, technology-driven solution for strengthening revenue assurance, optimizing inspection resource deployment, and advancing data-informed tax governance in India.