Analysis and Visualization of News Articles via Sparse Statistics

The researchers aim to provide web-based tools for the analysis and visualization of a large body of text documents, with a focus on databases of news articles. They are using advanced algorithms, drawing from recent progresses in machine learning and statistics, to allow a user to quickly produce a short summary and associated timeline showing how a certain topic is described in news media.

By starting with their current prototype, the scientists are developing a web-based service allowing users to analyze large and medium-sized databases (including some user-loaded data), and post comments on the results. This service could represent a first step towards a community-based service geared towards a real-time, easy-to-use critical analysis and commentary forum on news media. The design of our tool will be driven by a case study focusing on the image of various ethnic groups and religions in the national media.

Interactive Visualization