math.similarity: adding a few similarity metrics.

db4
John Benediktsson 2012-05-03 10:57:30 -07:00
parent a9a0080ce3
commit 6b8d802985
3 changed files with 34 additions and 0 deletions

View File

@ -0,0 +1 @@
John Benediktsson

View File

@ -0,0 +1,18 @@
! Copyright (C) 2012 John Benediktsson
! See http://factorcode.org/license.txt for BSD license
USING: math.functions math.similarity tools.test ;
IN: math.similarity.tests
CONSTANT: a { 1 2 1 5 1 0 0 }
CONSTANT: b { 0 0 0 0 2 3 1 }
{ t } [ a a euclidian-similarity 1.0 1e-10 ~ ] unit-test
{ t } [ a b euclidian-similarity 0.1336766024001917 1e-10 ~ ] unit-test
{ t } [ a a pearson-similarity 1.0 1e-10 ~ ] unit-test
{ t } [ a b pearson-similarity 0.2376861940759582 1e-10 ~ ] unit-test
{ t } [ a a cosine-similarity 1.0 1e-10 ~ ] unit-test
{ t } [ a b cosine-similarity 0.5472455591261534 1e-10 ~ ] unit-test

View File

@ -0,0 +1,15 @@
! Copyright (C) 2012 John Benediktsson
! See http://factorcode.org/license.txt for BSD license
USING: kernel math math.statistics math.vectors sequences ;
IN: math.similarity
: euclidian-similarity ( a b -- n )
v- norm 1 + recip ;
: pearson-similarity ( a b -- n )
over length 3 < [ 2drop 1.0 ] [ corr 0.5 * 0.5 + ] if ;
: cosine-similarity ( a b -- n )
[ v* sum ] [ [ norm ] bi@ * ] 2bi / 0.5 * 0.5 + ;