Natural Language Search with Knowledge Graphs - Trey Grainger, Lucidworks
Ғылым және технология
To optimally interpret most natural language queries, its important to understand a highly-nuanced, contextual interpretation of the domain-specific phrases, entities, commands, and relationships represented or implied within the search and within your domain.
Learn about Semantic Knowledge Graphs, how to build them, how to use them, how to combine them with entity extraction through Solr's Text Tagger, and how to build a search application that adequately interprets the nuanced meanings associated with natural language queries.
Speaker: Trey Grainger, Chief Algorithms Officer, Lucidworks
ACTIVATE Conference: www.activate-conf.com
Lucidworks: www.lucidworks.com
Пікірлер: 4
Great presentation with useful examples. Thanks Trey.
Awesome content. Thanks!
great presentation , the more i think about it, the more SKG sounds like word2vec , take average of word embedding for Java and Scala and find word embedding(s) that are close to it. Then if context make a difference use embedding generated by language models, this way we can differentiate between BBQ in 2 different contexts as well.