top_n_field_type.xml

<fieldType name="top10_title_vector_synonyms" class="solr.TextField" positionIncrementGap="100" autoGeneratePhraseQueries="false">
		<analyzer type="query">
			<charFilter class="solr.HTMLStripCharFilterFactory" />
			<!-- MUST HAVE comma's mapped to spaces in here - required by vector and queryboost parsers -->
			<charFilter class="solr.PatternReplaceCharFilterFactory" pattern="[;:\'\&quot;\*/\),\(\-\|]" replacement=" "/>
			<tokenizer class="solr.WhitespaceTokenizerFactory"/>
			<filter class="solr.TrimFilterFactory" />

			<filter class="solr.SynonymFilterFactory" synonyms="jobs_titles.txt" ignoreCase="true"  tokenizerFactory="solr.WhitespaceTokenizerFactory"/>
			
			<!-- filter to just tokens in the file above -->
			<filter class="solr.TypeTokenFilterFactory" types="synonym_types.txt" useWhitelist="true"/>
			
			<!-- expand to top 10 terms weighted by similarity, at query time only -->
			<!-- MAKE SURE ignorecase="false" !!!!!! --> 
			<filter class="solr.SynonymFilterFactory" synonyms="top10_title_synonyms.txt" ignoreCase="false" expand="true" />
			<!-- extract payloads -->
			<filter class="solr.DelimitedPayloadTokenFilterFactory" encoder="float" delimiter="|"/>
			<!-- maps payloads to term weights -->
			<filter class="org.dice.solrenhancements.tokenfilters.PayloadQueryBoostTokenFilterFactory"/>
			<filter class="solr.LengthFilterFactory" min="1" max="10000000" />
		</analyzer>
		
		<!-- Index time config much simpler - just extract title terms -->
		<analyzer type="index">
		
			<charFilter class="solr.HTMLStripCharFilterFactory" />
			<!-- MUST HAVE comma's mapped to spaces in here - required by vector and queryboost parsers -->
			<charFilter class="solr.PatternReplaceCharFilterFactory" pattern="[;:\'\&quot;\*/\),\(\-\|]" replacement=" "/>
			<tokenizer class="solr.WhitespaceTokenizerFactory"/>
			<filter class="solr.TrimFilterFactory" />

			<!-- filter to just tokens in the file above -->
			<filter class="solr.SynonymFilterFactory" synonyms="jobs_titles.txt" ignoreCase="true"  tokenizerFactory="solr.WhitespaceTokenizerFactory"/>
			
			<filter class="solr.TypeTokenFilterFactory" types="synonym_types.txt" useWhitelist="true"/>
			<filter class="solr.LengthFilterFactory" min="1" max="10000000" />
		</analyzer>
		<similarity class="org.dice.solrenhancements.similarity.PayloadOnlySimilarity"/>
</fieldType>