Building Large-Scale Multilingual Fuzzy Matching Framework

Sponsor talk

Go to NumFOCUS academy page.

Fuzzy matching becomes an increasingly complex problem when consolidating multiple data sources to answer real-time queries. In this talk, we describe a framework to handle large-scale data and multilingual queries. Our framework is composed of multiple parts: building multilingual vector representations, querying multiple databases, and scoring retrieved entities in real time.


Abdulrahman Althobaiti

Data Scientist @ Mozn Interested in Topological Data Analysis (TDA) and explainable ML