Forecasting and Geospatial Analysis at Scale: Practical applications of ML

Topic: Forecasting and Geospatial Analysis at Scale: Practical applications of machine learning and analytics at an Alaskan telecom carrier

Description: Providing Internet, TV and wireless service in Alaska poses unique challenges. Geographic scale and environmental challenges magnify the cost of providing service and require careful forecasting of customer behavior and network performance to anticipate capital investments. Data and wireless network traffic measurement results in billions of data points. We will discuss how we use a cloud-based Spark clusters from Databricks and Azure to continuously forecast time series. We will also discuss the challenge of presenting information regarding network performance in a geospatial format useful to engineers and service designers using Apache Sedona, PostGIS, various visualization tools.

Bio: Matt Beattie is currently the Director of Data Science for GCI, the leader provider of Internet, TV, and wireless services in Alaska. He is also an Adjunct Professor of Data Science and Analytics at the University of Oklahoma in Norman. Prior to joining GCI, Matt was the Head of Sales for the Western Region of Crown Castle, and held multiple leadership positions at the General Manager and VP level at AT&T and Cingular Wireless. Matt holds a Masters’ Degree in Data Science and Analytics from the University of Oklahoma, a Masters’ Degree in Operations Research from North Carolina State University, and a Bachelor of Science in Engineering from Duke University. Matt is pursuing his PhD in Engineering at the University of Oklahoma, where his work focuses on the application of machine learning to topics in epidemiology and addiction research.