Graph & Beyond Lunch Break #11: Fuzzy Search

When dealing with real-world text retrieval, we often not only care about exact matches to our search phrase but need to consider for example typos or alternative spellings.

“Fuzzy search” is an umbrella term referring to a set of algorithms for such approximate matching. Usually, such algorithms evaluate some similarity measure showing how close a search term is to the items in a dictionary. Then a search engine can make a decision on which results have to be shown first.

In this Lunch Session, we will take a look at different similarity measures and show how fuzzy search works in ArangoDB.

About the Presenter:

Simran is a data modeling and migration enthusiast with a cultural sciences background. He is responsible for ArangoDB’s documentation, helps out with design questions, and supports the community, especially with AQL and JavaScript.