Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{188621,
        author = {Tanya Raikwar and Neer Pandey and Nityasri kanukolanu and Gayithri N},
        title = {RainSafe: A Hybrid ML and Threshold-Based Framework for Hyperlocal Urban Flood Risk Assessment},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {7},
        pages = {3142-3147},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=188621},
        abstract = {Flooding is a serious and persistent problem that puts communities, economies, and environments at serious risk. For urban areas like Bengaluru, RainSafe is a real-time flood-risk monitoring and alerting system. The system incorporates real-time weather data, user-generated flood reports, and a Random Forest Classifier (RFC) based on machine learning (ML) to assess localized flood risk levels.
A novel hybrid risk engine ensures more dependable real-time decision-making by combining threshold-based rules with machine learning predictions. FastAPI and MongoDB are used in the backend's implementation to enable scalable data ingestion and alert generation. The design, methodology, and evaluation of RainSafe are described in this paper, emphasizing its capacity to deliver street-level risk assessment with high responsiveness and low computational cost. The system achieved an accuracy of 87% and F1-score of 0.84, exhibiting improved reliability},
        keywords = {Flood detection, threshold based alert algorithm (TBA), random forest classification (RFC), machine learning (ML), hybrid models, flood management, predictive analytics, urban hydrology.},
        month = {December},
        }