Projects‎ > ‎

DFG Crowdsourcing

Crowdsourcing is a newly emerging service platform and business model in the Internet. In contrast to outsourcing, where a job is performed by a designated worker or employee, crowdsourcing means to outsource a job to a large, anonymous crowd of workers in the form of an open call. This human crowd is abstracted by a crowdsourcing platform, which distributes the work submitted by an employer among the human worker resources and acts as mediator between worker and employer. The crowdsourcing paradigm is changing dramatically the future of work and work organization in the Internet [Hoß11]. .The goal of the research project is to design and to evaluate new mechanisms in order to optimize crowdsourcing as emerging paradigm for the work organization in the Internet. A successful crowdsourcing platform requires a robust and high-performance operation as well as a balance of interests among employer and employees. Therefore, novel mechanisms being developed in the project need to consider all stakeholders. On one hand, the mechanisms aim at reducing job completion times while improving the quality of work which is the key motivation for a long-term usage of crowdsourcing by the employer. On the other hand, a fair and convenient distribution of jobs among the workers in the crowd is a major requirement by the employer. In this context, an appropriate formulation of jobs and the splitting of work into smaller jobs is a key aspect. Another major goal aims at the integration of job- and worker-recommendation systems for employees and employers into crowdsourcing platforms, respectively..Beyond the design of concrete mechanisms, abstract models will be derived for a unified description of crowdsourcing platform and relevant use cases. The models allow to analyze the impact of parameter changes on the performance of platforms and quality of work. Hence, the models are used for evaluation and optimization of the mechanisms designed in the project. Further, the models help to develop a basic understanding of the interaction of platform mechanism and to predict future developments. The fundamental of the models are semantic and statistical analysis of existing crowdsourcing platforms, their workers, and jobs. The project questions the applicability of a unified model for different platforms and use cases or the need for partly specialized models. In particular, the following use cases will be considered in the project: Fraktur recognition and digitization of historical documents, subjective user studies on Quality of Experience, evaluation of recommendation systems for e-learning, mobile crowdsourcing considering device features and its location.

The project Crowdsourcing is funded by Deutsche Forschungsgemeinschaft (DFG) in the subject area "Operating, Communication and Information Systems" since 2012. "Design and Evaluation of new mechanisms for crowdsourcing as emerging paradigm for the organization of work in the Internet". The original German title of the project is "Design und Bewertung neuer Mechanismen für Crowdsourcing als neue Form der Arbeitsorganisation im Internet".