EVENT - Thursday
When6:00 PM - 9:30 PM
DSF Day 4 – Data Science for Game Analytics at King
Data Science Festival Day 4 – King (Ballot ticket only)
An evening at King. Best known for developing Candy Crush, the King platform supports 1.5 billion games a day, and 149 million daily active users.
Please register for a ballot ticket here: Get Tickets
Due to the popularity of Data Science Festival events, we are now allocating event tickets via a random ballot. Registering here enters you into the ticket ballot for the Data Science Festival Event at King on April 19th, 2018, the ballot will be drawn by the 16th April 2018. Those randomly selected will then be e-mailed tickets for the event, with the joining details.
If you get an allocated ticket, please bring a copy of your paper ticket or your ticket on your phone to the event to check in with your QR code. Tickets are non-transferable.
The Data Science Festival is the first of its kind as the only community-led, free to attend Data Science Festival in the UK.
6.00pm Guests arrive
6.30pm-Speaker 1: Jessica van der Kroef – How to spot biases in your datasets
7.00pm – Drinks food & networking
7.30pm – Speaker 2: Zhaozhi Qian – Marketing freemium Apps with the help of machine-learning
8.00pm – Speaker 3: Tom Matcham – Building an Automated Analytics System
8.30pm – 21:30pm – Networking
Speaker 1: Jessica van der Kroef – How to spot biases in your datasets
Summary: Jessica will be sharing her experience working for one of the biggest social games developers. The second part of her talk will be about a problem that she encountered recently in her work: selection bias. It’s a problem that is often not immediately clear, but has a tendency to creep up on us when we are not paying attention. Jessica will teach us about the different types of selection bias that we should watch out for, and show us how to spot them.
Bio: Jessica Van Der Kroef, a proud Dutchie with a passion for dance and music, moved to London in February 2015. She completed a bachelor’s and a master’s degree in econometrics from the University of Groningen, having written her master’s thesis at the genetics department of University College London. She started working for King as a data scientist intern, and is now a fully-fledged member of the Business Performance Unit in London. Working on live games like Farm Heroes Saga, but also on projects that are still in development, her work involves close interaction with the level designers, developers, and other members of the game team to ensure an optimal experience for our millions of players all over the world.
Speaker 2: Zhaozhi Qian
Talk Abstract: Marketing freemium Apps with the help of machine-learning. Zhaozhi will talk about two data science problems in the context of marketing: Churn Prediction and Customer Lifetime Value Prediction. Through these two examples, he will highlight the special considerations for marketing freemium products and some general thoughts on how to design production-ready machine-learning systems.
Bio: Since he joined King in 2016, Zhaozhi has been working on applying rigid scientific thinking to marketing, a field dominated by creativity and imagination. Zhaozhi spends lots of time understanding player behaviour, fighting off fraudsters, and making sure the campaigns are ROI positive. If Zhaozhi is not waiting for his hive query to finish or groaning about undocumented code, he must be cooking up the next system that streamlines marketers’ workflow and helps them make more informed decisions.
Speaker 3: Tom Matcham
Talk Abstract: The talk will be a deep dive on some of the technology Tom’s used to build a system that runs funnel and spend analysis without any prior knowledge of the data. There’s some novel data science being done using techniques which aren’t so popular that he thinks are really valuable.
Bio: Tom quit his mathematics phD in 2012 to make video games personalisation software and hasn’t looked back since. He’s a bit of an all-rounder, having built data warehousing solutions and worked as a games analyst but mainly considers himself a data scientist. Since late 2015, Tom has been working on an automated data scientist to help smaller companies access advanced analytical techniques. Tom now has a prototype working on live data which he is keen to show the World! In the little spare time he gets, Tom likes to moan about how terrible Python is, how slow R is and build his own data processing meta-programming language in C (yes, just plain C).