Join me in welcoming four great speakers: Ciaran Jetten, Marjolein Peters, Elien van Riet, and Django van Amstel. We’ll start and end the evening with some delicious food and beverages (can you guess what they’ll have on tap?). Finally, during our book raffle, you’ll have chance to win “Data Science from Scratch (2e)” and “TensorFlow for Deep Learning”.
- 5:30 PM: Walk-in with food and drinks
- 6:30 PM: Introduction by Jeroen Janssens (Data Science Workshops)
- 6:35 PM: Welcome by Ciaran Jetten (Heineken)
- 6:45 PM: Talk 1 by Marjolein Peters (CBS)
- 7:15 PM: Talk 2 by Elien van Riet (Reviewscan)
- 7:45 PM: Book Raffle
- 7:50 PM: Talk 3 by Django van Amstel (Heineken)
- 8:20 PM: Community Announcements
- 8:30 PM: Drinks
- 9:30 PM: End
Abstracts and Speaker Biographies
Talk 1: Innovation at Statistics Netherlands (CBS)
Statistics Netherlands (CBS) stands for a society which is well-informed on the basis of actual facts. As the national statistical office, we offer insights into societal developments by providing reliable statistical information and data. These insights support the public debate, academic research, policy development, and decision making processes. This has been our role since 1899… but innovation is essential to CBS. We are always looking for new ways of collecting data for our research. In doing so, we seek to publish with greater timeliness, detail and efficiency, without increasing the response burden by sending more surveys. In this presentation, I will discuss a number of big data initiatives we started. What techniques did we use? And what are the implications of these initiatives? Finally, I will show one method used in my own project about Dutch Hospital Care Data.
Marjolein is statistical research at Statistics Netherlands (CBS). She is working at the Department of Health & Welfare, focused on statistics of hospital and elderly care. Next to producing annual statistics, she is working on customized research requests by the Ministry of Health, Welfare & Sport. Prior to joining CBS, Marjolein obtained her PhD at Erasmus University Rotterdam, on the topic of complex genetics and human ageing. During her PhD, she visited one of the ENCODE data analysis centers (Lawrence Berkeley National Laboratory, California) for 3 months. Marjolein lives in Zevenhuizen with her husband and daughter, where she enjoys family life, running, playing tennis, and having dinners with friends.
Talk 2: API Demo Reviewscan: From Prototype to Product
Data teams know that a machine learning prototype in Jupyter Notebook is still far from being a deployed product. This summer, our team won the AI challenge of Rotterdam.ai with Reviewscan, a new product that unlocks data value of online reviews with AI. Since then we have moved beyond prototyping, validating our business proposition and building an automated pipeline using GCP. In this talk, I will explore what it takes to build a ML-product by giving a demo of the Reviewscan API. I will also discuss what challenges are faced when operationalising machine learning.
Elien innovates products and services by making sure data science meets the business case directly. She started to work on data science projects four years ago, first as a strategist for Amsterdam Smart City, then as experiment designer at Makerstreet and now as founding partner at Reviewscan. Elien is part of Women in AI network, a do-tank working towards gender-inclusive AI that benefits global society. She holds a Master in Humanities and is certified in Python for Data Science. In her free time she explores Amsterdam North where she enjoys family life, friends and great breweries.
Talk 3: Optimization of Order Allocations
Heineken Poland has hundreds of customer orders per day, with multiple warehouses across the country to potentially serve from. Facing variable stock levels and finite logistical capacities, solving the puzzle of where to allocate each order from is a difficult task currently solved by hand. Sub-optimal allocation results in inefficiencies such as unnecessary travelled distance by the delivery trucks, increase of rush-orders and out-of-stock events. In this talk, Django will introduce the traditional Allocation Problem from literature and show how the standard formulation was extended and applied to the Polish order allocation challenge.
Django van Amstel is a data scientist in the Insights Lab: the Competence Centre for Advanced Analytics of Heineken International. In this role, he develops and applies various modelling and optimization techniques to build data-driven products for the whole Heineken organization around the world. Before Heineken Django worked on high-tech machinery, building the actuators used in simulators for pilot training, equipment testing and entertainment. He has a background in Mechanical and Control engineering.