Hey HN'ers.<p>A couple of us are part-way through developing our next app (http://pageradius.com/), and the more we get into it, the deeper we're getting into machine learning and graph theory (including frequent itemsets, association rules, bayesian classification, clustering, and so on).<p>We've scoured the web for as much info as we can and are making pretty good progress, but really want to get stuck into a solid, well-edited book that explains these concepts clearly - we're really loving these subjects.<p>Any recommendations?<p>Thanks!
It isn't published yet, but you can get the early access (MEAP) of <i>Machine Learning in Action</i> from Manning:<p><a href="http://www.manning.com/pharrington/" rel="nofollow">http://www.manning.com/pharrington/</a><p>A good basic graph theory book is:<p><a href="http://www.amazon.com/Introductory-Graph-Theory-Gary-Chartrand/dp/0486247759/ref=sr_1_1?s=books&ie=UTF8&qid=1314381624&sr=1-1" rel="nofollow">http://www.amazon.com/Introductory-Graph-Theory-Gary-Chartra...</a><p>Depending on exactly what you're trying to do, you might also find some value in something like:<p><a href="http://www.amazon.com/Network-Science-Applications-Ted-Lewis/dp/0470331887/ref=sr_1_2?s=books&ie=UTF8&qid=1314381704&sr=1-2" rel="nofollow">http://www.amazon.com/Network-Science-Applications-Ted-Lewis...</a>
Disclaimer: I'm a total n00b on this topic.<p>I'm trying to learn how to classify items as related within a dataset. I know <a href="http://directedge.com" rel="nofollow">http://directedge.com</a> does this (funded by yc, run by #wheels) so I had a look at their articles which are a helpful beginners intro. <a href="http://directededge.com/tech.html" rel="nofollow">http://directededge.com/tech.html</a><p>In one of the articles #wheels recommends <a href="http://oreilly.com/catalog/9780596529321" rel="nofollow">http://oreilly.com/catalog/9780596529321</a><p>so I'm going to pick that one up but to be clear I really have no idea if this book addresses graph theory specifically; at this point anything and everything is helpful to me.
Hilary Mason's machine learning video is good:<p><a href="http://oreilly.com/catalog/0636920017493" rel="nofollow">http://oreilly.com/catalog/0636920017493</a><p>It's more of a primer, but it seems like most of the other material assumes that you're already a practitioner. It does have enough depth to actually be somewhat proficient after watching it.
The Elements of Statistical Learning is pretty good. It's also available for free.<p><a href="http://www-stat.stanford.edu/~tibs/ElemStatLearn/" rel="nofollow">http://www-stat.stanford.edu/~tibs/ElemStatLearn/</a>
I realise somebody asked a similar question about Graph Theory books already - <a href="http://news.ycombinator.com/item?id=865749" rel="nofollow">http://news.ycombinator.com/item?id=865749</a> - but that discussion is two years old, and I was hoping there'd be something newer by now!<p>Also, I've found this free online (downloadable) book on Graph Theory to be really useful: <a href="http://code.google.com/p/graph-theory-algorithms-book/" rel="nofollow">http://code.google.com/p/graph-theory-algorithms-book/</a>
For graph theory, I recommend Doug West's <i>Introduction to Graph Theory</i>. <a href="http://www.amazon.com/Introduction-Graph-Theory-Douglas-West/dp/0130144002" rel="nofollow">http://www.amazon.com/Introduction-Graph-Theory-Douglas-West...</a> It's somewhat expensive, but worth the price. Also, Doug is a cool dude.
Whats missing on the list is interesting works of Jon Kleinberg. Hes CS professor at CMU checkout this link <a href="http://www.cs.cornell.edu/home/kleinber/networks-book/" rel="nofollow">http://www.cs.cornell.edu/home/kleinber/networks-book/</a>