How can you tell if a patient is the same person across all the different electronic systems used in a hospital?
Can you be confident with messy data when lives are on the line?
Medical startup Luminare faced this challenge in a hospital setting and used Clojure to save the day and make the nurses happy again.
This talk will explore the challenge of record linking: dealing with dirty data sets, the pros and cons of different solution approaches, and using the Felligi-Sunter method to create a probabilistic algorithm to match records.
1) What is Record Linkage? ie: how to recognize the problem
2) What is the space of possible solutions? Different kinds of data require different solution approaches. How to recognize which methods will work best with your dataset.
3) Creating a probabilistic algorithm using the Felligi-Sunter method. I will walk the audience through the creation of a real-world probabilistic algorithm using Clojure.
Chris Oakman is a software developer, designer, and educator from Houston, TX.
He works at Luminare - a medical startup based out of the Texas Medical Center - and teaches software development at DigitalCrafts - a coding bootcamp school.
He is the author of several open source projects, including the cljs.info cheatsheet, the CLJS logo, and several Parinfer ports and editor plugins.
We will be meeting at The Cannon, which is nearby the intersection of Beltway 8 and I-10. While our normal meeting time is the third Wednesday, this month we will be meeting on Tuesday November 19, 2019.
Please use RSVP to this meeting here. RSVPs help us plan for seating and order the right amount of food for attendees.