Jesus Perez - Profile on Academia.edu (original) (raw)

Guido Zuccon related author profile picture

Carsten Eickhoff related author profile picture

Siamak Faridani related author profile picture

Iadh Ounis related author profile picture

Jorge Morato related author profile picture

Matt Lease related author profile picture

HyunJoon  Jung related author profile picture

Martin Davtyan related author profile picture

Thomas Hofmann related author profile picture

Mark Dredze related author profile picture

Uploads

Papers by Jesus Perez

Research paper thumbnail of University of glasgow (qirdcsuog) at TREC crowdsourcing 2011: TurkRank - Network-based worker ranking in crowdsourcing

NIST Special Publication, 2011

For TREC Crowdsourcing 2011 (Stage 2) we propose a networkbased approach for assigning an indicat... more For TREC Crowdsourcing 2011 (Stage 2) we propose a networkbased approach for assigning an indicative measure of worker trustworthiness in crowdsourced labelling tasks. Workers, the gold standard and worker/gold standard agreements are modelled as a network. For the purpose of worker trustworthiness assignment, a variant of the PageRank algorithm, named TurkRank, is used to adaptively combine evidence that suggests worker trustworthiness, i.e., agreement with other trustworthy co-workers and agreement with the gold standard. A single parameter controls the importance of co-worker agreement versus gold standard agreement. The TurkRank score calculated for each worker is incorporated with a worker-weighted mean label aggregation.

Research paper thumbnail of University of glasgow (qirdcsuog) at TREC crowdsourcing 2011: TurkRank - Network-based worker ranking in crowdsourcing

NIST Special Publication, 2011

For TREC Crowdsourcing 2011 (Stage 2) we propose a networkbased approach for assigning an indicat... more For TREC Crowdsourcing 2011 (Stage 2) we propose a networkbased approach for assigning an indicative measure of worker trustworthiness in crowdsourced labelling tasks. Workers, the gold standard and worker/gold standard agreements are modelled as a network. For the purpose of worker trustworthiness assignment, a variant of the PageRank algorithm, named TurkRank, is used to adaptively combine evidence that suggests worker trustworthiness, i.e., agreement with other trustworthy co-workers and agreement with the gold standard. A single parameter controls the importance of co-worker agreement versus gold standard agreement. The TurkRank score calculated for each worker is incorporated with a worker-weighted mean label aggregation.

Log In