This paper presents a framework for modeling and optimizing max algorithms commonly used in crowdsourcing. The model allows tuning the number of human responses needed and the size of item sets to be compared for achieving a certain level of quality given time and budget constraints.
D. Skoutas, and M. Alrifai. In Proc. of 20th ACM international conference on Information and knowledge management (CIKM '11), ACM, New York, NY, USA, 221-230., (2011)