The applet abbove shows the tree structure inherent in the segmentation of the search string, for values of k=1, 3, and 7, and also the apriori probability of the occurrence of a word based on a listing of the frequency of letters in the English alphabet (courtesy
Jeremy T. Teitelbaum). Simply choose a sample text, a value for k, and enter the string you wish to search for in the search box. Clicking the Search button will generate the tree representing the breakdown of the string(see
previous explanation for more details) and perform a search on the selected sample text. Note that only the approximate string matching for
k=1 is fully implemented, but carrying out a search with
k=3 or
k=7 can still be informative(some small searches are possible at this time). When each search is carried out, letters are all cast to lower case and the approximate string matches are determined using the selected value for
k to determine the allowed number of deletions, insertions and substitutions in a match.
All quotes found at
Dr. Gabriel Robins' page