Back

Speaker "Jan Neumann" Details Back

 

Topic

How Comcast uses Data Science and Machine Learning to improve the Customer Experience

Abstract

At Comcast we use data science and machine learning for many different applications which directly or indirectly impact the customer experience. For example, we use machine learning algorithms to make it easier for our customers to find the content they love via personalized content discovery algorithms and the voice remote on the X1 set top boxes. We also apply data science to optimize capacity planning, and predictive caching at the global network scale. In this talk I will give an overview of some of the recent work that combines machine learning with big data processing to improve both the visible and invisible aspects of the customer experience.

Profile

I am a manager in the Comcast Labs DC research group developing novel solutions for big data analysis, recommendation systems, intelligent image understanding and automatic content extraction from video. My specialties include: 1) Expertise in creating innovative solutions for large-scale machine learning problems (e.g. recommender systems, forecasting models), intelligent image and video search, and automated scene analysis of broadcast and surveillance videos, 2) Proficiency in designing and developing large-scale real-time computer vision and recommendation/data analysis software using object-oriented methodologies (C/C++, C#, Python, Java, Scala, and Matlab), and 3) Experience in managing research projects, interacting with customers, and working with teams of researchers and software developers