Dabbara Nithish Bargav Chowdary, Gogu Arjun, Dudekula Gouse Peera, Chinthamanu Aravind Kumar

DOI: https://doi.org/10.5281/zenodo.23205634

Residents who notice local hazards such as broken streetlights, open manholes, or waterlogging usually have only centralized complaint channels or general social networks to turn to, and neither connects the report with people who live nearby. This paper presents SAMANVI, a mobile platform for hyperlocal problem reporting and peer assistance that combines five mechanisms in one workflow. A PostGIS spatial query restricts every feed to a 1 km boundary at the database level. A TF-IDF and Naive Bayes classifier assigns each report a low, medium, or critical urgency. A weighted-average Community Trust Score is updated from feedback after each completed task. A four-digit arrival PIN prevents a task from being closed unless the helper and the requester meet in person, and a randomized offset of 10–20 m hides exact problem coordinates on the public c map. The system uses a React Native client, a Python Flask REST backend, and a PostgreSQL database of 15 tables with WebSocket messaging. Functional testing of six workflows, including phone and Aadhaar verification, problem posting, the 1 km boundary rule, urgency triage, and arrival-PIN verification, passed in every case. The contribution is an integrated, working design rather than a new learning algorithm, and classifier accuracy and load-test latency are not yet reported. The work shows that database level locality, lightweight triage, and in-person verification can be combined in one low-cost community platform.

community trust score, location-based services, PostGIS, spatial filtering, text classification

SAMANVI presents a location-aware approach to coordinating community assistance around local problems. It combines PostGIS spatial filtering, TF-IDF and Naive Bayes urgency triage, a weighted-average Community Trust Score, real-time messaging, arrival-PIN verification, and a coordinate offset in a three-tier mobile system. Functional tests of the authentication, reporting, spatial-filtering, triage, and assistance workflows passed. The contribution lies in 7 combining established techniques from location-based services, trust management, text classification, spatial databases, and location privacy into one hyperlocal workflow, not in a new algorithm.

[1] B. Wellman and B. Leighton, “Networks, Neighborhoods, and Communities: Approaches to the Study of the Community Question,” Urban Affairs Quarterly, 1979.
[2] F. Earls, J. Brooks-Gunn, S. Raudenbush, and R. Sampson, “Project on Human Development in Chicago Neighborhoods: Community Survey, 1994–1995,” ICPSR Data Holdings, 1999.
[3] S. Scellato, A. Noulas, R. Lambiotte, and C. Mascolo, “Socio Spatial Properties of Online Location-Based Social Networks,” in Proc. ICWSM, 2011.
[4] Z. Wang et al., “Discovering and Profiling Overlapping Communities in Location-Based Social Networks,” IEEE Trans. Syst., Man, Cybern. Syst., vol. 44, no. 4, pp. 499–509, 2014.
[5] M. J. Chorley, R. M. Whitaker, and S. M. Allen, “Personality and location-based social networks,” Computers in Human Behavior, vol. 46, pp. 45–56, 2015.
[6] Y. Ruan and A. Durresi, “A survey of trust management systems for online social communities: Trust modeling, trust inference and attacks,” Knowledge-Based Systems, vol. 106, pp. 150–163, 2016.
[7] M. Ahmadian, M. Ahmadi, and S. Ahmadian, “A reliable deep representation learning to improve trust-aware recommendation systems,” Expert Systems with Applications, 2022.
[8] A. Zarghami, S. Fazeli, N. Dokoohaki, and M. Matskin, “Social Trust-Aware Recommendation System: A T-Index Approach,” in Proc. WI-IAT Workshops, 2009.
[9] H. Allam et al., “Text Classification: How Machine Learning Is Revolutionizing Text Categorization,” Information, 2025.
[10] L. Dwarakanath et al., “Automated Machine Learning Approaches for Emergency Response and Coordination via Social Media in the Aftermath of a Disaster,” IEEE Access, vol. 9, pp. 68917–68931, 2021.
[11] J. Howard and S. Ruder, “Universal Language Model Fine tuning for Text Classification,” in Proc. ACL, 2018.
[12] L. Zhang and J. Yi, “Management methods of spatial data based on PostGIS,” in Proc. PACCS, 2010.
[13] PostgreSQL Global Development Group, “PostgreSQL: The World’s Most Advanced Open Source Relational Database,”
2023. [Online]. Available: https://postgresql.org
[14] R. Shokri, J. Freudiger, and J.-P. Hubaux, “A Unified Framework for Location Privacy,” in HotPETs Workshop, 2010.
[15] G. Sun et al., “Location Privacy Preservation for Mobile Users in Location-Based Services,” IEEE Access, vol. 7, pp. 87425 87438, 2019.
[16] B. Jia et al., “A Blockchain-Based Location Privacy Protection Incentive Mechanism in Crowd Sensing Networks,” Sensors, vol. 18, no. 11, p. 3894, 2018.
[17] A. Cullen and S. Kairam, “Practicing Moderation: Community Moderation as Reflective Practice,” in Proc. ACM HCI, 2022.
[18] Meta Platforms Inc., “React Native: Learn once, write anywhere,” 2024. [Online]. Available: https://reactnative.dev
[19] Pallets Projects, “Flask: A lightweight WSGI backend framework,” 2024. [Online]. Available: https://flask.palletsprojects.com
[20] Google LLC, “Google Maps Platform – Geocoding API,”
2023. [Online]. Available: https://developers.google.com/maps

Dabbara Nithish Bargav Chowdary, Gogu Arjun, Dudekula Gouse Peera, Chinthamanu Aravind Kumar (2026). SAMANVI: A Location-Aware Local Community Platform Using PostGIS-Based Spatial Filtering, NLP Urgency Triage, and Community Trust Scoring. International Journal of Computer Science Engineering Techniques, 10(5), 223–229. ISSN: 2455-135X. DOI: https://doi.org/10.5281/zenodo.23205634