A survey of task-oriented crowdsourcing

Since the advent of artificial intelligence, researchers have been trying to create machines that emulate human behaviour. Back in the 1960s however, Licklider (IRE Trans Hum Factors Electron 4-11, 1960) believed that machines and computers were just part of a scale in which computers were on one side and humans on the other (human computation). After almost a decade of active research into human computation and crowdsourcing, this paper presents a survey of crowdsourcing human computation systems, with the focus being on solving micro-tasks and complex tasks. An analysis of the current state of the art is performed from a technical standpoint, which includes a systematized description of the terminologies used by crowdsourcing platforms and the relationships between each term. Furthermore, the similarities between task-oriented crowdsourcing platforms are described and presented in a process diagram according to a proposed classification. Using this analysis as a stepping stone, this paper concludes with a discussion of challenges and possible future research directions.This work is part-funded by ERDF-European Regional Development Fund through the COMPETE Programme (Operational Programme for Competitiveness) and by National Funds through the FCT-Fundacao para a Ciencia e a Tecnologia (Portuguese Foundation for Science and Technology) within the Ph.D. Grant SFRH/BD/70302/2010 and by the Projects AAL4ALL (QREN11495), World Search (QREN 13852) and FCOMP-01-0124-FEDER-028980 (PTDC/EEI-SII/1386/2012). The authors also thank Jane Boardman for her assistance proof reading the document.