Hi everyone, I've been researching and relatively new to machine learning however familiar with distributed processing. I was wondering if there was a potential use case for mahout within the pub / sub model.
I've been considering intercepting all messages and if possible have the machine define features of the data, if data then begins arriving that is significantly outside the bounds of the current feature set then forcing a new event queue to be spawned in the parent system and publish the message to the new queue. Any thoughts on whether this would be a valid use case for mahout or suggestions? Thanks!
