Recommendation Systems for Digital Out of Home Advertising
Offered By: Toronto Machine Learning Series (TMLS) via YouTube
Course Description
Overview
Explore the world of recommendation systems for Digital Out of Home (DOOH) advertising in this 32-minute talk by Nataliya Portman, Senior Data Scientist at Cineplex Digital Media. Dive into the challenges and strategies of reaching the right audiences with financial products and services through networks of digital screens. Learn about a probabilistic modeling approach developed at CDM and how it drives decision-making for content selection and placement. Discover the differences between online and DOOH content, content categorization techniques, and data sources used. Gain insights into geographic audience representation, the workings of Flex SmartEngine, and the use of PowerBI for model insights. Understand the challenges of backend automation and methods for measuring recommended content performance in the DOOH advertising landscape.
Syllabus
Intro
DOOH EXAMPLE
THE IMPACT OF DOOH
WHAT IS DYNAMIC DOOH
DOOH HAS ITS CHALLENGES
HOW FLEX SMARTENGINE WORKS
THE CHALLENGE
WITH FLEX SMARTENGINE
THE CONTENT
ONLINE VS DOOH
CONTENT CATEGORIZATION
DATA SOURCES
GEOGRAPHIC REPRESENTATION OF AUDIENCE
PROBABILISTIC MODELING
POWERBI MODEL INSIGHTS DASHBOARD- DEMO
LIBRARY OF ADS REFERENCE TABLE FOR ADS COLLECTED OVER TIME
SMARTENGINE BACKEND AUTOMATION CHALLENGE RESOLUTION
HOW TO MEASURE RECOMMENDED CONTENT PERFORMANCE
SUMMARY
KEY TAKEAWAYS
Taught by
Toronto Machine Learning Series (TMLS)
Related Courses
Introduction to Artificial IntelligenceStanford University via Udacity Natural Language Processing
Columbia University via Coursera Probabilistic Graphical Models 1: Representation
Stanford University via Coursera Computer Vision: The Fundamentals
University of California, Berkeley via Coursera Learning from Data (Introductory Machine Learning course)
California Institute of Technology via Independent