Sign in
 
HomeLibraryWikiLearnGalleryDownloadsSupportForumsBlogs
Resources for IT Professionals >
> 
> 
> 
Query Normalisation

Answered Query Normalisation

  • Monday, April 23, 2012 8:51 PM
     
     
    Sign In to Vote
    0

    Hi guys, 

    Are full stops normalised to spaces?

    For example the searches "M.D.B" and "M D B" return the same results. Whilst, "MDB" returns different results.

    If there is a document titled "M.D.B and company" the following are the results:

    "MDB" returns the document first in the list.

    "M.D.B" and "M D B" will return the document late in the list.

    Is there a way to change this behaviour in FAST? 

    Emir

     

All Replies

  • Tuesday, April 24, 2012 8:42 AM
     
     
    Sign In to Vote
    0

    Hi

    this is as expected. because all the special characters are ignored at the time of indexing(dot,comma,asterick and other special character. the only way i can think  , you can noramalize the strings at the time of indexing by creating custom pipleline.

    Sriram S

     
  • Monday, April 30, 2012 2:39 PM
     
     
    Sign In to Vote
    0
    You can create synonyms to make MDB and M.D.B behave the same way.

    Dan Gøran Lunde

     
  • Sunday, May 20, 2012 7:13 PM
     
     
    Sign In to Vote
    0

    Hi guys,

    Thanks for the responses. Synonyms isn't really workable because their are hundreds of such entries and potentially more each week.

    Emir

     
  • Thursday, May 31, 2012 1:42 PM
     
     Answered
    Sign In to Vote
    0

    I ended up adding new managed properties and replacing special characters with binary strings at the pipeline processing stage.

    It is certainly not ideal but seems to work reliably.

    Emir

    • Marked As Answer by Emir123123 Thursday, May 31, 2012 1:42 PM
    •