> http://svnsearch.org/svnsearch/repos/ASF/search?path=/lucene/mahout
> Shouldn't be practice what we preach and mine our of statistics and predict, > cluster and classify :) Hmm.. I think Mahout looks very interesting :-) Seems to me it could be helpful in adding some features to SVNSearch I've been thinking about for a while: * Commit clusters: Identify which files are often touched in the same commit or in a certain window of time. * Developer clusters: Identify groups of people who typically work on the same parts of the code base. (SVNSeach kind of does this already, in a visual sense using graph layouts[1]) * Commit labelling: Is this a fix, feature, patch, refactor etc.. Do you guys think Mahout could be useful in developing featues like these? If anyone has ideas for other features, please do send me an email. I always enjoy feedback! Thanks, Eirik. (SVNSearch) [1] http://svnsearch.org/svnsearch/repos/ASF/search?view=graph&path=%2Flucene%2Fmahout
