Validating VGI Data Quality in Local Crowdsourced Accessibility Mapping Applications: a George Mason University Case Study




Rice, Rebecca M

Journal Title

Journal ISSN

Volume Title



As Volunteered Geographic Information (VGI) and other crowdsourcing techniques gain credibility as acceptable methods of data capture in situations where data is needed quickly, it is important to address the nature of uncertainty associated with these sources of data. This uncertainty includes the quality of data obtained from VGI contributors and the techniques used to ensure data quality. This thesis presents a method for assessing data quality in VGI applications based on the George Mason University Geocrowdsourcing Testbed, a web-based obstacle reporting application used to aid pedestrian navigation for the mobility and/or visually impaired. Quality assurance of geocrowdsourced data, especially the validation of position, is critical when the use is for routing handicapped pedestrians and wheelchair users. Implementation of the quality assessment methods includes the use of moderators to identify and fix errors in usergenerated data. This thesis addresses whether these techniques are consistent with