Hello,
until now, we use a stupid %like% SQL query script to assign the
following terms for Id/Item mapping in different id-spaces:
john wayne == john wayne
wayne, john == john wayne
I can imagine that Lucene offers much more possibilities for this
assignment.
Maybe with Lucene is also possible to match:
j. wayne == john wayne
john w. == john wayne
and do not link
J. Smith != john wayne
John S. != john wayne
I've searched the web on this topic and found a very old forum post that
it suggesting that lucene had not yet developed so far to solve this task.
http://mail-archives.apache.org/mod_mbox/lucene-java-user/200704.mbox/%3c84a3c846-73cb-4d04-aa7d-fd7d95780...@gmail.com%3E
I wonder whether there is now a clever standard approach using Lucene to
solve this problem?
I can imagine that a approach like the Spellchecker is using might work.
If the Spellindex holds all names of the second id-space and the first
id-space is used for the querrys.
String[] suggestions = spellchecker.suggestSimilar("john w.", 5);
But is there a better approach?
Can someone point me in the right direction for a effective approach?
Thanks in advance
Thomas
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org
For additional commands, e-mail: java-user-h...@lucene.apache.org
