CAREER: Computational Journalism: Integrating Algorithms and People in the Production of News Information

In this research I propose to develop a novel story discovery framework and sociotechnical system called NewsTips. NewsTips will serve as a testbed to understand how to effectively and efficiently combine experts, crowds, and algorithms in the discovery and conveyance of important public interest news leads to journalists. The research and development of NewsTips will focus on three diverse scenarios of news lead generation in order to ensure the generalizability of the framework developed. These scenarios include investigative journalism using online administrative documents, watchdog journalism related to factchecking claims in the media, and social journalism using online news comments. These diverse news information production scenarios set the context for the scientific challenges of the research: 1. Understanding the needs and difficulties of journalists with respect to semi-automated story discovery and news lead identification in different reporting scenarios; 2. Designing and developing (2a) Algorithmic and hybrid workflows (i.e. automated and human steps woven together) to identify and enrich newsworthy leads from large document corpora; and (2b) Information interfaces to present compelling news leads that are adaptable and controllable by users in order to fit varying contexts of use; 3. Evaluating workflows and information interfaces according to their utility and efficiency in different story discovery scenarios.