Visual analytics for an optimized and quality improved crowdsourcing process

From Master Projects
Jump to: navigation, search

has title::Visualization of disagreement-based quality metrics of crowdsourcing data
status: finished
Master: project within::AI and Communication
Student name: student name::Cristea Tatiana
Start start date:=2014/04/01
End end date:=2014/08/31
Supervisor: Lora Aroyo
Thesis: has thesis::Media:Thesis.pdf
Poster: has poster::Media:Posternaam.pdf

Signature supervisor



In this research we propose a visual-analytics model in the context of Crowd-Watson: a framework that uses Crowdsourcing methods to generate gold standard training and evaluation data for machine learning purposes. The proposed visual-analytic model aims to optimize the Crowdsourcing process and improve its quality by efficiently dealing with large and noisy amounts of data obtained as result of the Crowdsourcing process. Requirements for the dynamic, scalable and interactive visualizations were extracted from literature and through interviews with users of the framework.