Doug Turnbull - Staff Relevance Engineer, Shopify - Search as a constant experimentation cycle
Vector Podcast
00:00
The Importance of Tokenization in Search Engines
If I can engineer the similarity in the search engine so that it kind of uses the analysis to be like, oh, it has so many taxonomy nodes similar, that makes it more relevant. But maybe it has one or two dissimilar that makes a little less relevant. If I can sort of like zero in on a on that, then I'm really getting closer to whether it's like a stemmed version of this word or not. And you can create tokenization pipelines that take terms like let's say a myocardial infarction, which is a heart attack, and sort of like use a synonyms and other things to say,Oh, it's actually this part in this
Transcript
Play full episode
