Simon Hughes from Dice.com breaks down conceptual search at Solr/Lucene Revolution 2015. He goes into detail on how implementing conceptual search can boost recall and improve relevancy as well as solve several other issues. He also gives a brief demo of how Dice.com uses conceptual search with their jobs index and talks about the traditional techniques used for conceptual learning and their issues.
Presented at Lucene/Solr Revolution. Learn more: [ Ссылка ]
Implementing Conceptual Search in Solr
Теги
enterprise searchrelevancyfunctionalityrelevant matchesText SearchSoftwarebig dataconceptual searchApache SolrlucidworksapacheOpen Sourcelucenerelevant searchpresentationSolr LuceneRevolutiondice.comrelevancy tuningLucene/Solrsearch enginesDatalucid worksConventionsearchSolr RevolutionsolrrecallF1 Measurerelevantsemantic searchpolysemysynonymsconceptual learningword2vecvectorconceptual