Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
Author |
Topic |
aftalinc2009
Starting Member
1 Post |
Posted - 2009-04-02 : 12:31:57
|
Greetings, In context of SQL Server 2008's full-text search capability, does anyone know if there are any near future plans for supporting a true fuzzy search algorithm out of the box? We are evaluating SQL Server 2008’s full-text search capabilities and so far it has been really great with the exception of the Fuzzy search. So that the responder is aware of the length of my research here is what I already know: 1) I know Oracle uses the fuzzy question mark (?) operator to expand queries to include words that are spelled similarly to the specified term. This type of expansion is helpful for finding more accurate results when there are frequent misspellings in the documents in the database. This is the sort of capability we are looking to find in SQL Server 2008. 2) Currently the only fuzzy utility available via SQL Server 2008 (and prior versions) is an implementation of the "soundex" algorithm which is known for returning a lot of false positive and poor matches.3) I do know that Microsoft does have the newly acquired FAST ESP product whose fuzzy searching capabilities are considered to be phenomenal but my current interest is leveraging SQL Server’s full-text searching capability and determining if a true Fuzzy search support is in the cards. 4) I also do know that one can implement, using SQL CLR and a form of Ternary Search Tree, a flavor of Fuzzy searching algorithm within SQL Server 2008 but there is time and cost associated with this effort that we would like to avoid (see Pro Full-Text Search in SQL Server 2008 by Michael Coles with Hilary Cotter – which I have read cover to cover). So to sum up, I just want to know if, within the context of full-text searching, a true FUZZY search capability is in the works for SQL Server 2008 and beyond. If so, how soon ? Thanks, Tareen |
|
|
|
|