Last Thursday FUSE Labs UK launched Emporia, a social stream reader for users who want to "view the social web" through a lens of preferences, with a focus on Twitter. Think of it as providing a recommender system on top of the world of the social web.
Out of millions of tweets every day, Emporia uses the latest machine learning and statistical techniques to recommend a personal view based on your feedback. The more feedback you give, the more tailored the view you will have of the data.
It's already been a big hit, with lots of users. Large portions are implemented in F#, running on the Project Emporia's Azure-hosted servers in real time. Furthermore, FUSE Labs have taken the project form incubation to deployment in 9 weeks, a testament to the people involved, and great collaboration between Microsoft Research, FUSE Labs and Azure. Well done to Ralf, David, Jurgen, Allen, Ulrich, Shelley, Lili and the rest of the gang.