Solution Accelerator
Recommendation Engines for Personalization
Pre-built code, sample data and step-by-step instructions ready to go in a Databricks notebook
Increase conversion with personalized recommendations
Customers have different needs at each stage of the buyer journey. Choose the right recommender model for your scenario. Have a cold start problem? Try content-based recommenders. Nudging an existing customer to add to their cart? Wide-and-deep recommenders can help.
Image-based recommendations
Build a similarity-based image recommendation system for e-commerce that takes into account the visual similarity of items as an input for making product recommendations.
Get the notebooksMarket-based recommendations
Build a recommender that leverages product affinities to suggest additional items.
Get the notebooksWide-and-deep recommendations
Build a wide-and-deep recommender with collaborative filters that takes advantage of patterns of repeat purchases to suggest both previously purchased and related products.
Get the notebooksMatrix factorization (ALS) recommendations
Build a matrix factorization recommender to infer user ratings for various products. The alternating least squares (ALS) implementation for this recommender demonstrates patterns for matrix factorization that scale to accommodate the large numbers of user and product combinations found in real-world scenarios.
Get the notebooksCommon sense recommendations with LLMs
Use this Solution Accelerator to develop product recommendations based on common sense linkages for new-to-market products and optimized recommendation engines using large language models.
Resources
Blog
Blog
Blog
