PyData @ Imperva

We would like to thank Imperva for hosting us PHYSICALLY
Agenda
18:45-19:00 A word from our sponsor
19:00-19:30 ML with humans: Integrating experts into the learning process (Johnathan Azaria / Imperva)
19:30-19:45 A short break
19:45-20:15 Harnessing Data to Improve Healthcare / Moran Beladev
20:15-20:45 What's new in PyMongoArrow/ Pavel Duchovny

=======================
ML with humans: Integrating experts into the learning process

Human In The Loop (HITL) is a process in which, as part of the ML workflow, experts are asked their opinion about predictions made by an ML model in order to tune and improve the model. In this talk we’ll discuss how we created a mechanism to automatically predict the best security policies for our customers, and explain how we collaborated with and integrated engineers as a core part of our machine learning process, in order to both optimize the labeling process and quickly deploy to production without risking our customers. We’ll go through the different stages of the project, discuss the challenges we faced along the way and how we overcame them, and show how you can use a similar process for any heuristic/ML project you have.

==================
Harnessing Data to Improve Healthcare

In this talk, I will present how more than 60 billion historical medical visits were utilized to create an automated layer for digital healthcare. I will discuss NLP challenges, present our auto-tagging ML model and share our pipeline, which includes novelty deep models architecture built from scratch. I will also explain how Diagnostic Robotics formulates insights from claims data (a form of administrative data at a large scale that provides a great opportunity for AI in healthcare). You will understand how we use medical code embeddings and deep learning methods to build predictive proactive models that benefit the patients and reduce the cost of healthcare..

==================
What's new in PyMongoArrow

In this talk, I will cover an introduction to PyMongoArrow and its capabilities as well as share some of our upcoming roadmap on it for the python data community.