Hi, I think that`s not the way you can do it, because you cannot give a hint to your analyzer, which text fragment is more relevant than another at runtime. There is no marker so a filter process cannot know, which terms are to boost. You could write your own filter and let it read a file with some important terms in order to compare each term with your queryterms, but I think that would not be a good way.
If you have a way in order to split search query text into relevant terms. The first step is done. That`s a possible way for analysis at query time in order to search with right terms. In order to provide index data you can try to pre-process your data in order to save most important keywords in seperated search fields. Then you boost those fields on query time. Hope I could help, Dirk ----- erste Erfahrungen mit SOLR u. Vufind -- View this message in context: http://lucene.472066.n3.nabble.com/how-to-boost-query-term-after-tokenizer-tp4010889p4014245.html Sent from the Solr - User mailing list archive at Nabble.com.
