Web Content Recommendations using Bayesian Classification via Apache Mahout
Details
04-30 Update: The wifi code is wa0501. I believe the network is Cambridge. Pizza scheduled to arrive at 6:10.
Rani Nelken of Outbrain (http://www.outbrain.com/) has graciously offered to present on Bayesian Classification.
http://photos4.meetupstatic.com/photos/event/b/9/f/c/event_113987612.jpeg (http://www.outbrain.com/)
Rapid7 (http://www.rapid7.com/) is sponsoring pizza. Many thanks!!!
http://photos1.meetupstatic.com/photos/event/b/7/4/0/event_113986912.jpeg
The event will cover two main parts:
-
Bayesian Classification: a mathematical explanation will be given of this technique. One topic covered will be text classification as it pertains to content recommendation.
-
Software Demo: Apache Mahout is a software package that can be used for Bayesian Classification. Rani's demo will be similar to the following:
https://cwiki.apache.org/co (https://cwiki.apache.org/confluence/display/MAHOUT/Twenty+Newsgroups)nfluence/display/MAHOUT/Twenty+Newsgroups (https://cwiki.apache.org/confluence/display/MAHOUT/Twenty+Newsgroups)
Rani has sent over a link to an improved version of the above example.
https://github.com/nelken/Mahout-example/ (https://github.com/nelken/Mahout-example/blob/master/classify-20newsgroups.sh)
Here is a just another link that speaks to MatLab and Octave which is the software used in the Stanford Online ML course.
http://people.csail.mit.edu/jrennie/20Newsgroups/
https://www.coursera.org/course/ml
*. An additional short Mahout Tutorial is looking less likely for this event. Though if you are interested in helping with the development of a tutorial for a future workshop, then please let me know. Also, please note that the Hadoop group looks to have Ted Dunning as a speaker soon.
https://www.meetup.com/bostonhadoop/
Please note the talk will be on the 10th floor; the entrance is on the 11th floor, and then down a staircase. This is the setup as of Monday night for whatever event is happening. About 100 seats in the front section (seats on the sides too), and the staircase doubles as bleachers if needed. Tech events tend to range between 60% and 90% for %shows, 150*75% = 112.5.
http://photos4.meetupstatic.com/photos/event/4/a/1/c/event_115398972.jpeg