Time 
S 
Nick 
Message 
11:02 

pdurbin 
I have to admit my mind was a little blown listening to Grant Ingersoll describe how Lucene implements the http://en.wikipedia.org/wiki/Vector_space_model at http://www.seradio.net/2012/07/episode187grantingersollonthesolrsearchengine/ 
11:03 

pdurbin 
he was all "if you remember your trigonometry the cosine of [something] is [something]" ... vector this, vector that 
11:13 


aditsu joined #sourcefu 
11:14 

pdurbin 
"VSM score of document d for query q is the Cosine Similarity of the weighted query vectors V(q) and V(d)"  http://lucene.apache.org/core/4_2_1/core/org/apache/lucene/search/similarities/TFIDFSimilarity.html 
11:15 

pdurbin 
from "Classic Scoring Formula: Formula of Lucene's classic Vector Space implementation" at http://lucene.apache.org/core/4_2_1/ 
13:31 

pdurbin 
I'm listening to http://ocw.mit.edu/courses/mathematics/1806linearalgebraspring2010/videolectures/lecture9independencebasisanddimension/ but it's over my head. need to start with lecture 1... or (more likely) start with an easier course, a prereq 
15:57 

pdurbin 
done. nice lecture. it's always nice to hear the language of a domain, even if I'm completely confused. "null space" "matrix" etc. 
15:58 

pdurbin 
and I'm realizing I'm much more interested in [what I would consider] advanced mathematical concepts if I understand the practical application 
15:58 

pdurbin 
such as applying the vector space model to searching documents with Lucene 
16:01 

* pdurbin 
asks if this stuff is on topic in #pdl: http://irclog.perlgeek.de/pdl/20130504#i_7018368 
17:27 

pdurbin 
hearing about http://www.codeclub.org.uk at http://creativecodingpodcast.com/269/ 
17:27 

pdurbin 
"742 Code Clubs in the UK" 
17:42 


aditsu joined #sourcefu 