Errors
Information
About ONDA
For the past few years, newspapers have been moving their content online; the demand for free, high-quality content has increased and more people are getting their daily dose of news from RSS feeds. A huge number of RSS readers and aggregators have been developed in this time. They all do a great job of fetching the news, some of them do a great job of grouping together articles by topic; but there remains an interesting field yet to be investigated, with a social agenda: topical diversity.
ONDA stands for Online News Diversity Analyser. The system was developed as a dissertation project by Daniel Hough, a student of Software Engineering at the University of Sheffield. It uses machine learning techniques to 'cluster' together articles fetched from a number of newspapers' RSS feeds (see the imposed limitations) by their topic, and build up statistics about the topical diversity of the newspapers and present these statistics in cool graphical ways.
I hope you enjoy messing about with ONDA. If you encounter any glaringly obvious bugs, or even a little one which is getting on your nerves, check out the contact page.
When it's complete, you'll be able to download a PDF of the final dissertation report, including how the system was made and references.
Credits
Paul Ridgway very generously hosts ONDA at the moment, puts up with the regular hits to his CPU and is a constant source of support. Thanks Paul!
ONDA wouldn't be possible without the following third-party technologies
