[R] AGREP

Michael Friendly friendly at yorku.ca
Sat Feb 14 00:18:39 CET 2004


The problem of calculating levenshtein distances between strings reminds 
me of what I faced
years ago when writing an APL system to control interactive memory 
experiments, where subjects
typed words they could remember from a given list. 

To handle typos, I used the generalized outer product operator, null dot 
equals to create
a binary matrix comparing an input string with each correct word,

input o.= word

giving a 0/1 array whose rows reprepresented the letters in the input 
and whose columns
the letters in the output, from which varous measures of string 
similarity could be calculated
using reduction operators, or even a correlation, based on the positions 
of the 1s.
Thinking in arrays always helps avoid explicit loops.

-- 
Michael Friendly     Email: friendly at yorku.ca 
Professor, Psychology Dept.
York University      Voice: 416 736-5115 x66249 Fax: 416 736-5814
4700 Keele Street    http://www.math.yorku.ca/SCS/friendly.html
Toronto, ONT  M3J 1P3 CANADA




More information about the R-help mailing list