Archive for the ‘Alogrithms’ Category

Statistics on the similarity algorithms

July 27, 2010

So, just under a month ago I posted the following on similarity algorithms.

I have found it quite interesting to see which have been viewed, and the percentage share. Initially Damerau Levenshtein was the hot favourite, but then N-gram started coming through the ranks and is a clear leader with a 39% share of the hits!

I thought I would put together a simple histogram of the percentage share of hit for each similarity posting:

Disappointed my favourite, Markov chains, is languishing so far back…