Reuters Tracer: Toward Automated News Production Using Large Scale Social Media Data

摘要

To deal with the sheer volume of information and gain competitive advantage,the news industry has started to explore and invest in news automation. In thispaper, we present Reuters Tracer, a system that automates end-to-end newsproduction using Twitter data. It is capable of detecting, classifying,annotating, and disseminating news in real time for Reuters journalists withoutmanual intervention. In contrast to other similar systems, Tracer is topic anddomain agnostic. It has a bottom-up approach to news detection, and does notrely on a predefined set of sources or subjects. Instead, it identifiesemerging conversations from 12+ million tweets per day and selects those thatare news-like. Then, it contextualizes each story by adding a summary and atopic to it, estimating its newsworthiness, veracity, novelty, and scope, andgeotags it. Designing algorithms to generate news that meets the standards ofReuters journalists in accuracy and timeliness is quite challenging. But Traceris able to achieve competitive precision, recall, timeliness, and veracity onnews detection and delivery. In this paper, we reveal our key algorithm designsand evaluations that helped us achieve this goal, and lessons learned along theway.