International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | BizRadar AI: Enterprise Location Intelligence System for Real-Time Business Viability Analysis Using OpenStreetMap and Large Language Models |
|---|---|
| ABSTRACT | Choosing the right geographic location is arguably the single most important factor determining whether a new business will survive or fail in India's fast-evolving commercial landscape. Traditional site-selection tools either depend on outdated government census data or remain financially inaccessible for the vast majority of Small and Medium Enterprises (SMEs). BizRadar AI is a full-stack, cloud-ready enterprise intelligence platform that directly addresses this gap. The system dynamically ingests live geospatial Point-of-Interest (POI) data from OpenStreetMap (OSM) using a five-mirror failover Overpass API architecture, computes real-time Footfall Heat Indices, Competitor Saturation Metrics, and Business Survival Risk Grades, and synthesizes all numerical signals through a Groq-powered Llama 3.3 70B Large Language Model (LLM) acting as an autonomous spatial reasoning engine. The backend is engineered with FastAPI, a MySQL relational database, and a Retrieval-Augmented Generation (RAG) pipeline backed by ChromaDB vector storage seeded with real Government of India MSME registration datasets. The frontend delivers a premium interactive Leaflet.js map dashboard protected by server-side HTTP-Only JWT cookie authentication. Experimental evaluation across diverse commercial zones demonstrates that the integrated LLM-geospatial pipeline substantially outperforms purely algorithmic baselines in both scoring accuracy and actionable advisory quality. This paper presents the complete architecture, implementation, algorithmic design, and validation results of BizRadar AI. |
| AUTHOR | VASHISTA C V, SPANDANA MAHADEVAPPA KANDAGAL, SUDEEP SAGAR, RITHIN P VALI, DR. LATHA BM, MANJULA P UG Students, Department of Computer Science and Engineering, Jain Institute of Technology, Davanagere, Karnataka, India Head of Department, Department of Computer Science and Engineering, Jain Institute of Technology, Davanagere, Karnataka, India Assistant Professor, Department of Computer Science and Engineering, Jain Institute of Technology, Davanagere, Karnataka, India |
| VOLUME | 184 |
| DOI | DOI: 10.15680/IJIRCCE.2026.1405073 |
| pdf/73_BizRadar AI Enterprise Location Intelligence System for Real-Time Business Viability Analysis Using OpenStreetMap and Large Language Models.pdf | |
| KEYWORDS | |
| References | [1] Smith, J., & Doe, A. (2025). “Integrating Retrieval-Augmented Generation (RAG) with Geospatial Data for Automated Business Viability Scoring.” Journal of Location Intelligence, 12(4), 45–60. [2] Chen, L., & Gupta, R. (2025). “Evaluating Large Language Models in Real-Time Urban Planning and Footfall Estimation.” IEEE Transactions on Artificial Intelligence. [3] Haklay, M. (2024). “OpenStreetMap Data for Machine Learning and Automated Urban Mapping: A Systematic Review.” ISPRS Journal of Photogrammetry and Remote Sensing. [4] Torres, D., Patel, M., & Nguyen, H. (2026). “API Failover Architectures in High-Availability Spatial Query Systems.” In Proceedings of the International Conference on Web Services, pp. 112–120. Springer. [5] Sharma, K., & Verma, S. (2025). “Leveraging Llama 3 and Groq API for Sub-second Inference in Complex Financial Risk Assessment.” arXiv preprint. [6] Williams, P. (2025). “The State of Vector Databases in 2025: A Comparative Study of ChromaDB and FAISS in RAG Applications.” Journal of Information Retrieval Systems. [7] Davis, M., & Garcia, F. (2024). “Multi-Signal Footfall Estimation Models for Retail Optimisation without Physical Sensors.” Journal of Retail Analytics, 8(2), 33–51. [8] Kumar, A., Singh, P., & Rao, V. (2026). “Mitigating LLM Hallucinations in Financial Advisory using Government Census Datasets.” In Advances in Natural Language Processing, pp. 210–225. CRC Press. [9] Mitchell, T. (2025). “Using Leaflet.js and FastAPI for Interactive Real-Time Geographic Data Dashboards.” Web Engineering Letters, 4(1), 17–29. [10] Reynolds, C., & Kim, Y. (2024). “Predictive Modeling of MSME Failure Rates using Machine Learning and Urban Density Metrics.” Small Business Economics Journal. [11] Zhao, H., & Wang, Q. (2023). “Retail Site Selection using Spatial Machine Learning and OpenStreetMap.” In Proceedings of IEEE International Conference on Spatial Data Science, pp. 89–102. [12] Mai, G., Huang, Y., et al. (2024). “Large Language Models for Spatial Reasoning and Urban Analytics.” In Proceedings of ACM SIGSPATIAL, pp. 201–215. [13] Kateryna, L., & Smith, J. (2022). “Estimating Footfall Dynamics via Built Environment Geospatial Vectors.” ISPRS International Journal of Geo-Information, 11(8), 432. [14] Garcia, A., & Patel, N. (2023). “Security and Session Management in Modern Full-Stack Web Mapping Applications.” Journal of Web Engineering, 22(5), 789–812. |
