Newdle: Interactive Visual Exploration of Large Online News Collections

Abstract

Newdle is a visual-analytics system for exploring tagged online news collections. Newdle automatically constructs an article network and article-tag network for a collection and conducts clustering and path analyses on them. Using the analysis results, the system builds visualizations and supports user interaction with the networks. A topic overview lets users grasp a large collection's contents at a glance. It intuitively displays significant article clusters' semantics and temporal features in the collection. Users can also conduct in-depth analyses on topics, tags, and articles. A working prototype of Newdle uses New York Times RSS feeds as its example data input. Case studies with this prototype illustrate Newdle's effectiveness and efficiency.