The second workshop on Crowd Science is organized in conjunction with the 47th International Conference on Very Large Data Bases (VLDB 2021). This workshop is the second in a series of events that has the goal of helping crowdsourcing “transition” from art to science, and tackles the research challenges that we face to make crowdsourcing a technology that users can easily access and leverage, and produces results that researchers and businesses can rely on. In addition to regular paper submissions, this year we have organized a crowdsourced audio transcription shared task that attracted 18 participants around the world. Eight submissions have beaten a strong baseline method, thus advancing the state-of-the-art in this challenging task. This workshop features three invited talks, seven paper presentations, an overview of this shared task, as well as a panel discussion. We received 11 submissions, out of which 7 were accepted as the talks at the workshop after peer review. Besides the accepted papers, the volume contains a shared task overview by its organizers. We thank all the authors for their contributions and the efort they put into it, and we are very grateful for the excellent work of our reviewers and program committee members. Last but not least, we would like to thank VLDB organizers for their assistance that made the workshop organization process a pleasure. The 2021 edition of the workshop took place in a hybrid format, and we look forward to the forthcoming (in-person) editions of the Crowd Science workshop.
• Anjana Arunkumar, Arizona State University • Marcos Baez, University of Trento • Alessandro Bozzon, Delft University of Technology • Fabio Casati, ServiceNow • Geert-Jan Houben, Delft University of Technology • Alexey Kushnir, Carnegie Mellon University • Nikita Pavlichenko, Yandex • Sihang Qiu, Delft University of Technology • Ivan Stelmakh, Carnegie Mellon University • Andrea Tocchetti, Politecnico of Milano • Dmitry Ustalov, Yandex • Jie Yang, Delft University of Technology • Xiong Zhou, Amazon