top of page
OUR SPEAKERS
Share on:
Tal Laron is a Data Science Team Lead at Riskified, responsible for Riskified’s models and DS components. Tal has extensive data science experience across multiple fields, and, before Riskified, she worked in several startups. Tal holds a B.Sc in statistics and economics from the Tel Aviv University and a master’s degree in financial economics, a global program from the Reichman University. As a former gymnast she enjoys sports and sporting events, beside her great love for data and sports, she is a mother of two, Assaf & Noa.
Tal Laron
Data Science Team Lead @ Riskified
English, Hebrew
Languages:
Location:
Herzlyia, Israel
Can also give an online talk/webinar
Paid only. Contact speaker for pricing!
MY TALKS
Build Your Own Lego Model: Connecting Historical Data to Improve Models’ Accuracy
Data / AI / ML, Leadership
My 5 year old son is crazy about Lego. He taught me there are multiple ways to connect two Lego bricks. This made me realize that this approach can be applied to my work as a data scientist
Using Lego, I will explore how different data points can be connected and combined to build a robust model for improving fraud detection in e-commerce using machine learning, which is what we do at Riskified.
For example, addresses can be connected and combined by geographical distance or by textual similarity, which gives us a hybrid approach for improving our models.
By the end of this talk, you will gain an understanding of how hybrid techniques can improve the performance of predictive models and build your own hybrid lego model.
Build Your Own Lego Model: Connecting Historical Data to Improve Models’ Accuracy
Completed
true
Visible
true
Order
4
bottom of page