Crowdsourced Data Management: A Survey

摘要: Any important data management and analytics tasks cannot be completely addressed by automated processes. These tasks, such as entity resolution, sentiment analysis, image recognition can enhanced through the use of human cognitive ability. Crowdsouring platforms are an effective way to harness capabilities people (i.e., crowd) apply computation for tasks. Thus, crowdsourced has become area increasing interest in research industry. We identify three problems management. (1) Quality Control: Workers may return noisy or incorrect results so techniques required achieve high quality; (2) Cost The crowd is not free, cost control aims reduce monetary cost; (3) Latency workers slow, particularly compared computing time scales, latency-control required. There been significant work addressing these factors designing developing manipulation operators, optimizing plans consisting multiple operators. In this paper, we survey synthesize a wide spectrum existing studies on Based analysis then outline key that need considered improve