期刊名称：Bulletin of the Technical Committee on Data Engineering
出版年度：2019
卷号：42
期号：4
页码：8-22
出版社：IEEE Computer Society
摘要：The goal of this article is to propose an optimization framework by acknowledging human factors toenable label acquisition through active learning . In particular, we are interested to investigate tasks,such as, providing (collecting or acquiring) and validating labels, or comparing data using activelearning techniques. Our basic approach is to take a set of existing active learning techniques for a fewwell known supervised and unsupervised algorithms, but study them in the context of crowdsourcing,especially considering worker-centric optimization (i,e., human factors). Our innovation lies in designingoptimization functions that appropriately capture these two fundamental yet complementary facets,performing systematic investigation to understand the complexity of such optimization problems, anddesigning efficient solutions with theoretical guarantees..
