Open AccessApplication and improvement of capture-recapture model for crowdsourced testing
Author(s) -
Yi Yao,
Yuchan Liu
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1738/1/012125
Subject(s) - crowdsourcing , computer science , task (project management) , process (computing) , machine learning , data mining , engineering , systems engineering , world wide web , operating system
Crowdsourced testing has become an effective alternative to traditional testing. However, crowdsourced testing is inherently difficult to manage. In this paper, objective information such as defect count is considered to evaluate the task process. We researched how the capture-recapture model estimates the total number of defects in a crowdsourced testing scenario. By comparing the difference between crowdsourced testing and traditional testing, incremental sampling methodology and the best estimator for crowdsourced testing were proposed to estimate the number of defects, so as to improve the ability to evaluate the completion degree of crowdsourced testing task.