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{193190,
        author = {Hemavathi J and Dr.Rama satish KV and Lovely Sasidharan},
        title = {An Intelligent Spatio-Temporal Deep Learning Framework For Urban Crowd Flow Prediction},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {4264-4267},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=193190},
        abstract = {Rapid urbanization, increasing population density, and growing dependence on mobile connectivity have intensified crowd congestion in modern cities, particularly in public spaces such as transportation hubs, commercial centers, educational campuses, and tourist destinations. Existing crowd monitoring systems primarily rely on fixed infrastructure such as CCTV cameras or manual observation, which are costly, difficult to scale, and often limited to real-time analysis without predictive capabilities. These limitations hinder effective crowd management, safety planning, and mobility optimization. This project presents CrowdMind, an AI-powered crowd detection and forecasting system that leverages anonymized GPS-based mobility data to generate intelligent insights into crowd behavior. The system employs DBSCAN clustering to identify real-time crowd density hotspots and uses Long Short-Term Memory (LSTM) networks along with Prophet time-series models to predict future crowd patterns based on historical and temporal trends. A modular backend architecture integrates data ingestion, preprocessing, machine learning, storage, and application services to ensure scalability, performance, and reliability. CrowdMind provides a user-friendly mobile interface that visualizes crowd density through heatmaps, predictive graphs, and navigation guidance, enabling proactive decision-making for users and authorities. The system emphasizes data privacy by processing anonymized location data while supporting real-time operations and high user concurrency. Experimental evaluation demonstrates high accuracy, responsiveness, and robustness under varying data loads. By combining real-time crowd detection with predictive analytics, CrowdMind bridges the gap between raw mobility data and actionable urban intelligence, supporting applications in smart city planning, transportation management, public safety, tourism management, and sustainable urban development.},
        keywords = {CrowdMind- Crowd flow Prediction},
        month = {February},
        }