Time Series with Python
Kaizen Data Conference, San Francisco, September 16th, 2016
Abstract:
Machine Learning is used by companies like Kickback to analyze video games data to provide users with the best experience. One of its most interesting applications is catching players who use cheats and bots to defeat their opponents. Time series analysis is at the core of this, since a video game is essentially a sequence of actions in time. In this workshop you will learn the caveats you need to keep in mind when applying standard machine learning techniques to Time Series, and you will design your first algorithm to catch a cheater.
Speaker:
Francesco Mosconi is data scientist at Catalit LLC. He has briefly served as Head of Data Science at Kickback, a company enabling money bets on skill-based videogames and was previously Chief Data Officer at Spire, a company that invented the first wearable respiration tracker. 3 times founder, YCombinator and Singularity University graduate, he earned a joint PhD in biophysics at University of Padua and Université de Paris VI and has a master degree in theoretical physics. Francesco runs weekend workshops called DataWeekends, where you can learn the fundamentals of machine learning and deep learning. Find out more at www.dataweekends.com.
Requirements:
- python
- pandas
- numpy
- scikit-learn
Environment:
Install Anaconda Python 2.7