Automatically Discovering Unknown Product Attributes Impacting Consumer Preferences – MSI

Working Paper

Automatically Discovering Unknown Product Attributes Impacting Consumer Preferences

Ankit Sisodia

Yale Universiy

Alex Burnap

Yale University

Vineet Kumar

Yale University

Mar 28, 2022

Analyzes unstructured data (images) using advanced machine learning techniques for disentangled representation learning to derive attributes of consumer preference without human intervention.

Topics

Consumer Behavior & Insights marketing analytics Social Media & Social Influencers

research recap

Finding Unknown Attributes of Consumer Preference

Marketing models analyze structured product attributes to determine consumer preferences. Deriving such attributes from images, text and other unstructured data typically requires human judgment and can be hard to scale. Using available marketing data, machine learning techniques for disentangled representation learning can derive meaningful attributes without human intervention, often revealing consumer decision factors missed by other approaches.

By using MSI.org you agree to our use of cookies as identifiers and for other features of the site as described in our Privacy Policy.