Crowd knowledge creation plays a central role in many types of Web based information systems, ranging from community question-answering (CQA) systems (e.g. StackOverflow and Quora) to micro-task crowdsourcing systems (e.g. Amazon mTurk and CrowdFlower). In these systems, knowledge demands are generally fulfilled by means of tasks (e.g. questions in CQA systems, micro-tasks in crowdsourcing systems) executed by group of individuals (e.g. contributors in CQA systems, workers in crowdsourcing systems). Despite of the success in some platforms, knowledge creation tasks so far are assumed to be of low cognitive complexity and are generally solved as a bottom-up process; as a consequence, outcomes are heavily dependent on the spontaneous and autonomous contribution of crowds. This limits our ability to control the volume, speed, and quality of knowledge creation. By unlocking the value of features related to human knowledge, e.g. expertise and motivation, we envision that crowd knowledge creation can reach its full potential where complex, cognitively intensive tasks are solved and thus high-quality knowledge is efficiently generated...
|Award date||15 Nov 2017|
|Publication status||Published - 2017|
Bibliographical noteSIKS Dissertation Series No. 2017-47
- Knowledge Creation
- Human Computation
- Recommender Systems
- User Modeling