Talk 1: Class Imbalance in Fraud Detection
Speaker: Mor Sondak
Abstract: Many real world machine learning problems need to deal with imbalanced class distribution i.e. when one class has significantly higher/lower representation than the other classes. Often performance on the minority class is more crucial, like in our industry, fraud detection.
In this talk we will discuss several techniques to handle class imbalance in classification like resampling and weighted loss function. We will present our greedy tree algorithm that overcomes the highly imbalanced dataset.
Bio: Mor works as lead Data Scientist at Fraugster (https://fraugster.com/). The team at Fraugster uses AI to eliminate online payment fraud, to reduce consumer frustration and increase businesses’ growth. Mor works on the Machine Learning engine that classifies the transactions in real time. Before Fraugster she worked on Machine Learning based recommendation systems for e-commerce. She holds a Master of Science (MSc) focused on Computer and Information Sciences from the Technion - Israel's Institute of Technology, with research in Information Retrieval.
Talk 2: TBD
Are you organizing Class Imbalance in Fraud Detection; more?
Claim the event and start manage its content.I am the organizer