Few more things that could be highlighted Dirichlet Process clustering(the first model based clustering implementation), Co-occurrence based recommender(the first pure hadoop recommender?), Document Vectorizer (its mixed with LLR, so might have to highlight the high quality ngram vectors they generate?) FP-Bonsai is more like an improvement of PFPGrowth not a separate implementation. So could be written as: Speed boost in FPGrowth using FP-Bonsai pruning. Introduce Colt's new home as Mahout math. Colt is pretty famous.
On Mon, Mar 15, 2010 at 7:15 PM, Sean Owen <sro...@gmail.com> wrote: > Here is the release announcement I checked in: > > <section> > <title>18 March 2010 - Apache Mahout 0.3 released</title> > <p>The Apache Lucene project is pleased to announce the > release of Apache Mahout 0.3.</p> > <p> > Highlights include: > <ul> > <li>New math, collections modules</li> > <li>LLR colocation implementation</li> > <li>FP-bonsai implementation</li> > <li>Hadoop-based Lanczos SVD solver</li> > <li>Shell scripts for easier running of algorithms, > examples</li> > <li>... and much much more: code cleanup, many bug fixes > and performance improvements</li> > </ul> > </p> > <p>Details on what's included can be found in the<a > href=" > https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310751&styleName=Html&version=12314281 > "> > release notes</a>. > </p> > <p>Downloads are available from the > <a href="http://www.apache.org/dyn/closer.cgi/lucene/mahout/ > ">Apache > Mirrors</a> > </p> > </section> > > On Mon, Mar 15, 2010 at 1:42 PM, Grant Ingersoll <gsing...@apache.org> > wrote: > > +1. > > > > I verified all signatures/MD5. Ran some samples that I have. > > > > Looks good. Do we have a Release Announcement draft somewhere? Lots of > good stuff in this release. > > > > -Grant >
