CHS: Small: Data-Driven Retention in Crowdsourced Image Analysis and Mapping

本研究旨在提高个人对公民科学图像分析的参与度、维持度和重复参与度。提高对公民科学的参与度可能会加快问题解决和发现的步伐，允许更多人做出贡献，扩大参与，并为原本可能没有参与的社区提供新的参与途径。该项目力求通过将参与者与提出项目的组织联系起来，并鼓励参与者重新参与和协作，来服务和建立对区域复原力至关重要的较小的区域非营利组织。它将开发和部署一个开源网络平台和工具包，用于创建引人入胜的公民科学项目。推动应用的是灾害应对，随着灾害的频率和严重程度的增加，它有可能产生更广泛的影响。然而，一般的方法可以更广泛地应用于公民科学、众包、人类计算和人机交互社区。本项目有三个主要研究目标。首先，它将开发技术来自动改进在线公民科学项目的入职过程。这将有助于理解早期任务对长期参与这类项目的影响。其次，它的目标是开发一种参与者脱离模型，该模型可用于指导干预措施，以动态调整以适应个别参与者。这些干预措施将在多种任务类型、有限的参与者历史记录和工作流程中的相关转移的背景下进行审查。第三，虽然大多数众包公民科学项目是由正式的学术研究人员和大型组织推动的，但该项目的一个目标是建立一个研究平台，让中小型社区组织可以制定研究议程，探索他们追求的不同问题和目标。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This research aims to improve individual participation, retention, and repeat engagement in citizen science image analysis. Improving engagement in citizen science may accelerate the pace of problem solving and discovery by allowing contributions from a larger group of people, broaden participation, and allow new pathways for involvement for communities who may not otherwise have had them. This project seeks to serve and build smaller regional non-profit organizations that are vital to regional resilience by connecting participants with organizations proposing projects and encouraging participant re-engagement and collaboration. It will develop and deploy an open-source web platform and toolkit for creating engaging citizen science projects. The motivating application is disaster response, which has the potential for broader impact as disasters are increasing in frequency and severity. However, the general approaches can be applied more broadly in the citizen science, crowdsourcing, human computation, and human-computer-interaction communities.This project has three main research aims. First, it will develop techniques to automatically improve the onboarding process of online citizen science projects. This will help to understand the impact of early tasks on longer-term engagement in such projects. Second, it aims to develop a model of participant disengagement that can be used to guide interventions that dynamically adjust to individual participants. These interventions will be examined in the context of multiple tasks types, limited participant history, and relevant diversions within the workflow. Third, while most crowdsourced citizen science projects are driven by formal academic researchers and large-scale organizations, a goal of this project is to build a research platform where small to medium scale community based organizations can set the research agenda and explore the different questions and aims they pursue.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.