Ranjit JangamExecutive Director of Data Science, Comcast
Machine learning and data science plays significant role at cable & telecommunication industries. There is vast amount of data with 3600 view to develop models to drive revenue & growth, manage risk, improve customer experience, provide personalized experiences to consumers, developing innovative & reliable product experiences. Due to excessive competition in the industry, emphasis on data science driven machine learning algorithms including parametric or non-parametric, big data and AI tools has been severe and never been most important. Objective of this session is to review complete end to end cycle of developing machine learning models at scale including feature engineering, model development, model deployment, data & model governance and review real-world business applications of predictive models in the cable industry.
Rama MahajanamDirector of Machine Learning, Comcast
With the COVID-19 pandemic starting mid-March, 2020, the rapid development of an “outside network check” provided an opportunity for the ML/AI modeling team to gather better features and labels. With a keen desire to keep technicians and customers isolated, the team explored new ML/AI models. These new models are trained to use cloud-based RF measurements. These measurements include remote telemetry from DOCSIS, and other equipment logged by collection systems. Another set of measurements were taken at the tap and ground block, finally offering a way to segment the network and train the machines differently. We will review a few highlights of this fascinating exercise, which is currently underway.
The Institute for Operations Research and the Management Sciences
phone 1 443-757-3500
phone 2 800-4INFORMS (800-446-3676)