Newdle: Interactive Visual Exploration of Large Online News Collections

In this paper, we present a novel visual analytics system named Newdle with a focus on exploring large online news collections when the semantics of the individual news articles have already been tagged. Newdle automatically conducts clustering and relation analyses on news articles and builds visualizations and supports interactions upon these analyses. By providing a novel topic overview in which the semantics and temporal features of the significant article clusters in a large collection are intuitively displayed, Newdle allows users to grasp the content of the collection in a glance. Through the rich set of interactions and visualizations provided by Newdle, users can effectively conduct in-depth analyses on topics, tags, and articles of interest. We have implemented a fully working prototype of Newdle, using the online New York Times RSS feeds as its example data input. We present several case studies to illustrate the effectiveness and efficiency of Newdle.

• Type: Journal
• Year: 2010
• Where: CGA
• Authors: Jing Yang, Dongning Luo, Yujie Liu
• Keywords: Analyses, CGA 2010, Computational Geometry, Example Data Input, Significant Article Clusters