Local knowledge matters for crowdsourcing systems: Experience from transferring an American site to China

Abstract

We present the results of a preliminary cross-cultural study of Cyclopath, a geographic crowdsourcing system for cyclists. We created a Chinese version and tested it in lab with 40 cyclists in Dalian, China. We analyzed the data entered by users and found some significant differences compared to similar studies done in the US; this illustrates the need for local knowledge in crowdsourcing applications. From usage patterns and interview data, we observed that landmarks play a larger role for navigating the map and requesting routes; this suggests that integrating landmarks better into the Cy-clopath UI could improve Chinese users' experience with the system. Finally, we found that concepts of authority and hierarchy may influence Chinese users' trust of peer-contributed content; this suggests that reputation mechanisms may help improve trust. Copyright © 2013, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.