Friday, May 08, 2009

randomNode : revisited

After a >2 year delay since the original randomNode clustering search engine experimental project (and a related paper), I have finally decided to give it a shot and try to apply the concept in the case of practical real-world search engine. It seems that such approximate graph partitioning might be even more useful in the context of realtime data clustering (for example, for dynamic sources like twitter). Stay tuned for a upcoming demo .... :)


Post a Comment

<< Home