P00-1037 minor changes . For computing the Damerau-Levenshtein distance between two strings , this can
P11-1094 all spelling errors which were a Damerau-Levenshtein distance of 1 ( Damerau , 1964 ; Levenshtein
W13-2410 these " words " is computed using Damerau-Levenshtein distance ( Damerau , 1964 ) . Basically
W13-2410 alignment through POS tagging and Damerau-Levenshtein distance </title> Andrey Abstract The
W15-1901 distribution of cosine and normalized Damerau-Levenshtein distances ( nDL ) to target hypotheses
W15-1904 most probably using evaluation of Damerau-Levenshtein distance ( Damerau , 1964 ) between words
W13-2410 20 percent of total dataset ) . Damerau-Levenshtein distances for all sentences were computed
W13-2410 are processed as " words " and Damerau-Levenshtein distance between them is computed . This
W14-0105 correct , while the pairs 2The Damerau-Levenshtein distance which takes into account transpositions
J11-2005 version of what is now known as the Damerau-Levenshtein distance , a metric of the similarity
W13-2410 the al - gorithm , we found that Damerau-Levenshtein distance shows some kind of dis-balance
P00-1037 models have all been based on Damerau-Levenshtein distance measures ( Damerau 1964 ; Levenshtein
W13-2400 alignment through POS tagging Damerau-Levenshtein distance Andrey Kutuzov Ivelina Stoyanova
W13-2410 that there is correlation between Damerau-Levenshtein distance and the quality of alignment
W15-1901 used throughout this section : Damerau-Levenshtein distance ( nDL ) , a variant of Levenshtein
W13-2410 normalized " distance , DL is original Damerau-Levenshtein distance , sP is " POS watermark " for
D09-1039 We then compare them using the Damerau-Levenshtein distance Sdl ( dc , dys ) ( Dam - erau
P00-1037 previous models based on weighted Damerau-Levenshtein distance . Note that neither P ( f | ph
P06-1129 direction ever evolves from simple Damerau-Levenshtein distance ( Damerau , 1964 ; Levenshtein
W13-2410 characterized by normalized POS Damerau-Levenshtein distance of enormous 2.6 , however , human
hide detail