Journalists rely on image and video verification to support their investigations and often utilize open source tools to verify user generated content, but current practice requires experts be involved in every step of the process. Additionally, lacking custom tools to support verification efforts, experts are often limited to the utility of existing, openly available tools, which may or may not support the interactions and information gathering they require. We aim to support the process of geolocating images and videos through crowdsourcing. By enabling crowd workers to participate in the geolocation process, we can provide investigative journalists with efficient and complete verification of image locations. Parallelizing searching speeds up the verification process as well as provides a more extensive search, all while allowing the expert to follow up on other leads or investigative work. We produced a software prototype called GroundTruth which enables crowd workers to support investigative journalists in the geolocation of visual media quickly and accurately. Additionally, this work contributes experimental results demonstrating how the crowd can be utilized to support complex sensemaking tasks. / Master of Science
| Identifer | oai:union.ndltd.org:VTETD/oai:vtechworks.lib.vt.edu:10919/78697 |
| Date | 10 August 2017 |
| Creators | Kohler, Rachel |
| Contributors | Computer Science, Luther, Kurt, North, Christopher L., Horning, Michael A. |
| Publisher | Virginia Tech |
| Source Sets | Virginia Tech Theses and Dissertation |
| Detected Language | English |
| Type | Thesis |
| Format | ETD, application/pdf |
| Rights | In Copyright, http://rightsstatements.org/vocab/InC/1.0/ |
Page generated in 0.002 seconds