On Wed, Apr 30, 2014 at 1:29 PM, neha sinha <nehasinha...@gmail.com> wrote: > <filter class="solr.EdgeNGramFilterFactory" minGramSize="3" > maxGramSize="15" side="front"/> > <filter class="solr.SnowballPorterFilterFactory" language="English" > protected="protwords.txt" />
I think combining NGrams with Porter filters, especially in that order will do really weird things. Have you tried using the Admin console? You really want to see what happens to different words when you run them through your pipelines. Probably with debug mode enabled to see what effect NGram filter does for positions as well. Oh, and if you modified your index chain, did you reindex completely? You must, otherwise you have old processed tokens lying around. On the other hand, you can experiment with filter definition and not reindex (only reload core) until you see the text flowing through and being indexed/queries correctly. You are quite far from the normal scenario with your setup, so you are unlikely to get a magic answer, more like the pointers towards the tools that solve the problem. Regards, Alex. Personal website: http://www.outerthoughts.com/ Current project: http://www.solr-start.com/ - Accelerating your Solr proficiency
