Got feedback, questions, suggestions, bugs to report? Use this form:
Newstracker is a mashup application that combines news RSS feeds, web search and semantic web services. The application loads news feeds, extracts meaningful terms out of the feed items, and then queries search engines for web pages related to the news. The application also keeps track of which items you have seen and visualizes new items that were found since your last visit. Newstracker refreshes the feeds and searches the web every two hours.
The frontpage displays a list of recent items in a news feed. The available feeds are listed in the tabs on the top of the picture. A small star under the title of a feed item indicates that the feed item is new or changed since your last visit. Clicking the [in detail] link takes you to a page that shows the feed item and the related news, blogs and discussion pages in more detail. The [focus on] section lists meaningful terms associated with the news item. If you hover the mouse over a term, the context in which the term appears is shown in a tooltip. Clicking on a term shows other feed items that are also associated with the term.
For each news feed item, the center column lists top three related news, blog and discussion pages. Clicking on the small + button opens a preview of the related items. To see more related news, blog and discussion pages, click on the [in detail] link in the feed item.
On the top of the frontpage, a list of top terms extracted from recent feed items is displayed. Clicking on a term shows feed items associated with the term. Clicking on the Catch Up button marks all items as read. It works by saving a small cookie on your computer that stores the time of your last visit. After this, the star will appear on new feed and related items that you have not seen before.
Newstracker uses a number of external APIs (Application Programming Interface) to collect information from the web. First, the application loads RSS news feeds from MSNBC.com. It then extracts terms to search for from the news feed items. This is done through a combination of a proprietary algorithm and the ClearForest web service. Meaningful terms returned by the Clearforest service are also displayed along with the feed items and in the top tags section. Once the search terms have been extracted from the feed items, the application proceeds to query the MSN Search web service to retrieve news and discussion web pages related to the feed items. Technorati API is used to search for related blogs. The feed items, meaningful terms and the search results are stored in a database from which the various pages are generated.