Crowdsourced social media monitoring system development.

Kumar, Amit; Ross, Catherine L.; Karner, Alex A.; Katyal, Rohan · 2017 · ROSA P / Georgia. Department of Transportation. Office of Research

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This report addresses the gap between the state of the art in crowdsourced traffic management and current practice, specifically for the Georgia Department of Transportation (GDOT). The research was motivated by the limitations of traditional Traffic Management Center (TMC) technologies, such as CCTV cameras and loop detectors, which are costly to maintain and insufficient for monitoring all network links due to staffing constraints. Crowdsourcing offers a low-cost method to engage citizens as sensors, improving incident detection and public participation. However, challenges regarding data reliability, high volumes of unstructured information, and user engagement hinder widespread adoption. The study aims to evaluate crowdsourcing options and propose a tailored solution for GDOT. The methodology combined a comprehensive literature review with empirical data collection. The authors categorized crowdsourcing into active (user-generated content), passive (automatic sensor data), and combined models, analyzing their respective advantages and disadvantages. To understand practical implementation challenges, the team conducted interviews with TMC personnel across multiple states, including Utah, Florida, Michigan, and Washington D.C., focusing on their experiences with crowdsourced tools like Waze and Strava Metro. Additionally, the researchers performed an on-site assessment of GDOT’s TMC operations in Atlanta, conducting internal interviews and a SWOT analysis to identify specific workflow inefficiencies and technological needs. Key findings identified data reliability and the filtering of high-volume information as the primary obstacles for TMCs. The study revealed that while many DOTs utilize social media for dissemination, few have integrated crowdsourced data into automated incident detection workflows. Manual verification of crowdsourced reports creates a bottleneck that negates the efficiency gains of crowdsourcing. The report highlights that existing systems often suffer from low user retention and bias. Based on these findings, the authors proposed a low-cost, integrated system for GDOT consisting of a mobile application for citizen incident reporting and a text mining application that leverages Twitter’s infrastructure to automatically fetch and process reports. This architecture aims to minimize development costs while providing real-time data. The significance of this work lies in its actionable recommendations for modernizing TMC operations. The authors conclude that GDOT should adopt an integrated Traffic and Incident Management (TIM) architecture that combines multiple data sources to enable automated incident detection and validation. Specific recommendations include implementing computer vision to automatically detect incidents from camera feeds, utilizing social media analytics to filter data, and employing gamification to sustain user engagement. The report argues that investing in these crowdsourcing technologies is critical for GDOT to handle increasing data volumes and reduce operator burden, positioning the agency to benefit from the broader revolution in big data and artificial intelligence within transportation management.

Key finding

The proposed crowdsourced system utilizes a mobile application and text mining algorithm to automatically fetch, filter, and prioritize incident reports from Twitter, thereby reducing the manual workload for traffic management center operators.

Methodology

mixed_methods

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• incidence prevalence