Web 2.0 personalized news sites
To some people, news sites such as Reuters, CNN, NYTimes and Washington Posts seem so “yesterday”. Editors actively decide what news are important and what headline to run on the front pages, and on the other hand, readers passively receives this information. In Web 2.0, personalized news sites do things in a complete reverse. Readers get to decide what articles are important to them, and what headline articles should be ran in the front page.
A recent review showed that Web 2.0 personalized news sites are becoming popular. For a site to be classified as a personalized news site, it typically implements the following 3-steps information flow process:
Some key players in the fields are digg, reddit, spotback, Feeds2.0, LeapTag and Findory. Most of these sites exploit machine learning techniques to provide news recommendation service. Digg and reddit are leaders in the market, according to current Alexa traffic reports.
I very much like personalized news sites, but they also lack a feature that is important to a news-geek like me. Because the current personalized news sites don’t understand location information about their readers, they are unable to recommend news that is related to the local environment of the readers.
While news aggregation sites like Google News has dedicated pages for regional news, but its feature is not as “socialable” as which of digg and reddit. It will be great if future personalized news sites can recommend local news such as local police reports, traffic accidents, high-school events, county fairs and free concerts at a local park. Even better, the recommendation may also include blog posts from the local neighborhood people.
