String matching, distance between two strings. Works particularly well to detect retweets or tweet variations.

levenshtein.R

### string matching


### metric to find the similarity between two strings
### some context in:
### http://en.wikipedia.org/wiki/String_metric


### testing levenshtein metric
library(RecordLinkage)


str1 <- "scientific american"
str2 <- "american scientific"

levenshteinDist(str1, str2) # pure distance (number of changes needed to go from str1 to str2)
levenshteinSim(str1, str2) # similarity metric based on the distance metric