SIGIR
2011
ACM
13 years 2 months ago
2011
ACM
We study how to best use crowdsourced relevance judgments learning to rank [1, 7]. We integrate two lines of prior work: unreliable crowd-based binary annotation for binary classi...