A profile-aware microtasking approach for improving task assignment in crowdsourcing services

Abstract

Microtasking as one of the crowdsourcing models has found penetration mostly in developed nations and is slowly making inroads in developing nations. It has been used for quickly solving a myriad of social and business challenges by tapping into the diversity of the unknown crowd. Nevertheless, there are still a number of research challenges that are found in various microtasking initiatives, such as lack of user profiling, quality of task submissions, poor task design and assignment, ambiguous task assessments, and lack of diverse platforms to cater for the needs of differently skilled crowds. Most prominent microtasking platforms do not consider or even exploit the profile of the crowd in improving task design, assignment, and eventually quality of completed microtasks. In this paper, the objective was to explore the design and evaluation of a conceptual approach that could be implemented in microtasking environments for purposes of improving task assignments and eventually quality. A design science research methodology was adopted to build the proposed artefact, which was evaluated for relevance using comparative analysis and crowdsourcing metrics. The results suggest that exploiting micro workers' profiles could improve task design, assignment, assessment, and ultimately quality. However, stakeholders' awareness of effective microtasking approaches still requires research attention.