Just a thought. "scalable machine learning and data-mining libraries" ?. FPgrowth is not machine learning. Similary LDA is not machine learning but more like data modelling. I know, its all fuzzy, and wish we had a better way to say it. "tools for understanding patterns from data and predicting from learned ones" Many people who are not in this field wont know the difference even if we say its all machine learning.
On Tue, May 11, 2010 at 12:28 PM, Sean Owen <[email protected]> wrote: > No problem, I'll reword that accordingly. > > On Tue, May 11, 2010 at 3:16 AM, Ted Dunning <[email protected]> > wrote: > > I think that generally looks good except that I disagree with the > scalable = > > hadoop-based assertion. I think that scalable means that we handle big > > problems that are difficult with other tools. Hadoop is one key method > for > > this and we use it a lot. In other cases we use other methods. One > example > > is how taste handles delivery of realtime recs. Another is where clever > > algorithms like Pegasos or sgd which give large scale results on single > > machines. >
