Leveraging Data Insights From Risk Management to Marketing Optimization

It's been a while since the PyData Amsterdam 2024 conference, so it's time for another meetup! JetBrains Datalore and PyData Amsterdam are pleased to announce our first joint meetup. We have prepared several intriguing and practical talks from Adyen, ABN AMRO, and JetBrains. They will surely attract many involved in data analytics and LLMs. After the presentations, we will happily discuss these and other topics during the networking session. We invite everyone working in the fields of data science and data analysis. Hope to see you on October 24 at 18:00 at The Social Hub Amsterdam City, Wibautstraat 129, 1091 GL, Amsterdam.

SCHEDULE

• 18:00-18:45: Welcome with food and drinks! (🍕 / 🍺)
• 18:45-19:15: Talk 1 - "MPLX - Machine Learning Driven Risk Management" (Rogier Verlinden, Jesse Koreman - Adyen)
• 19:15-19:45: Talk 2 - "Optimize your marketing expenses: An overview of different methods" (Bas Stinenbosch - ABN AMRO)
• 19:45-20:00: Break
• 20:00-20:35: Talk 3 - "Lies, damned lies and large language models" (Dr. Jodie Burchell - JetBrains)
• 20:35-22:00: Networking / drinks!

TALKS
[Talk 1]: “MPLX - Machine Learning Driven Risk Management” by Rogier Verlinden & Jesse Koreman
Adyen processed more than 900 billion euro’s in 2023. With some of these transactions becoming a chargeback or refund, we need a comprehensive and adaptive risk system to mitigate our financial exposure. In this talk we’ll explain what crucial role is played by machine learning in this system, how we ensure scalability and what lessons were learned along the way

[Talk 2]: “Optimize your marketing expenses: An overview of different methods" by Bas Stinenbosch
How effective is your marketing? This is a question that not a lot of companies are able to answer. However, with the right data science methods you will be able save money and boost your sales. In this talk I will give an overview of the different methods to gain insights and show you how to use them. So, if you want to learn more about Multi Touch Attribution modeling with Markov Chains or Marketing Mix Modelling with a Bayesian approach, then this talk is the one for you!

[Talk 3]: “Lies, damned lies and large language models" by Dr. Jodie Burchell
Would you like to use large language models (LLMs) in your own project, but are troubled by their tendency to frequently “hallucinate”, or produce incorrect information? Have you ever wondered if there was a way to easily measure an LLM’s hallucination rate, and compare this against other models? And would you like to learn how to help LLMs produce more accurate information? In this talk, we’ll have a look at some of the main reasons that hallucinations occur in LLMs, and then focus on how we can measure one specific type of hallucination: the tendency of models to regurgitate misinformation that they have learned from their training data. We’ll explore how we can easily measure this type of hallucination in LLMs and end by looking at recent initiatives to reduce hallucinations in LLMs, using a technique called retrieval augmented generation (RAG). We’ll look at how and why RAG makes LLMs less likely to hallucinate, and how this can help make these models more reliable and usable in a range of contexts.