Social network mining of requester communities in crowdsourcing markets

摘要: Crowdsourcing is a new computing approach where human tasks are outsourced to large number of workers. has not only attracted attention from industry but also various academic communities. Amazon Mechanical Turk (AMT) been the first commercial platform offering crowdsourcing services its customers. AMT often referred as supplying ‘artificial’ artificial-intelligence. Recent research efforts have addressing analysis community structure large-scale platforms. In this work, we discuss detailed statistics popular marketplace provide insights in task properties and requester behavior. Here present model automatically infer communities based on keywords. Hierarchical clustering used identify relations between keywords associated with tasks. We novel techniques rank requesters by using graph-based algorithm. Furthermore, introduce models methods for discovery relevant brokers who able act intermediaries platforms such AMT.