Modeling challenges in fraud detection & Adapting Data Analytics to Change

Modeling challenges in fraud detection

Training machine learning models for fraud detection entails various challenges, such as data quality, biases, and data drift. This presentation addresses issues from different stages of the model training process, and the strategies to mitigate them: including advanced sampling techniques, feature engineering, and validation methods. Adri and Gábor will present real-world examples and key insights gained from developing fraud detection models at Mastercard.

Adrienn Juhasz, Lead Data Scientist at Ekata,Mastercard specializes in building machine learning models for fraud detection. In the last 5 years she was working on statistical models to identify online fraudsters using advanced techniques. She now leads the data science research team in Budapest, focusing on model development and fraud prevention strategies.

Gabor Nguyen is the Data Science Director at Ekata, a Mastercard company. He has managed multiple teams specializing in data science, ML Ops, and software engineering. Currently, he leads Personal Data Science and ML Operations at Mastercard Identity. His focus includes various data challenges in fraud prevention, such as data flows, data quality, governance, and modeling.

Adapting Data Analytics to Change: Lessons from Prezi’s Post-COVID Evolution

The post-COVID landscape brought significant shifts in product strategy, company organization, and the role of data teams at Prezi. In this talk, we’ll explore how our data analyst team adapted to these changes, balancing evolving business needs with the challenges of the legacy data stack built over a decade. I’ll share insights into our journey—how we adapted to the changing organization, tackled outdated infrastructure, aligned with new strategic goals in the AI era.

Balázs Szakács, Senior Director of Data Science and Analytics at [Prezi.com](https://prezi.com/).

Balázs has been working with data for 20 years and has lead teams for 15 years in companies like Telenor, Ustream, IBM, Graphisoft and Prezi. He is passionate about how data teams can make the biggest business impact and how to move on from legacy systems.

The agenda:
17:45 gates open
17:45-18:15 Warmup and chit-chat
18:15-19:30 Talks
19:30 - 21:00 Drinks, snacks, networking
By attending this event you agree to be photographed. Let us know if you would not like to appear on these pictures. We may use the pictures in social media.