bloom filters are used for approx matches, however, if you want the nearest word finding, and a dictionary is available, may be a trie, then till there is a match move, into the trie down, however if there is a mismatch, calculate the levenstein distan with the rest of the characters in the pattern and the down strings within the trie and list down the lowst distance words
On Jul 4, 8:40 am, souravsain <[email protected]> wrote: > Hi All > > I am looking for an algorithm for approximate string matching problem. > This is a common known problem and I do have one from Steven S Skiena > in his book Algorithm Design Manual [Chapter 8]. Link / guide to any > other good algorithm will be of great help.[Please don't send results > of google search. Looking for some source where member has read and > used / liked it] > > For those who are not familiar to this problem, it gives me a list of > possible words from known dictionary of words, if I type type a word > with some spelling mistake. Like I type "peple" and it should give me > "people","peble","purple" etc..... > > Thanks, > Sourav -- You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/algogeeks?hl=en.
