Solution Accelerator

Fuzzy Item Matching

Pre-built code, sample data and step-by-step instructions ready to go in a Databricks Notebook

Fuzzy item matching is an essential function in many retail and consumer goods organizations. Whether it's comparing new product offerings to ones already offered on a vast online marketplace to minimize seller redundancy, the scraping of competitor information on a website for price comparisons, supplier verification of online listings to ensure terms and conditions for sale are being met, or the harmonization of retailer and market analyst data with internal product hierarchies, many organizations spend a tremendous amount of time and energy identifying product matches between data sets in order to enable their work.

Optimize product matching to drive sales

Use machine learning and the Databricks Lakehouse Platform for product matching that can be utilized by marketplaces and suppliers for various purposes. Resolve differences between product definitions and descriptions and determine which items are likely pairs and which are distinct across disparate data sets.

Download notebook

Use images and metadata for fuzzy matching with Zingg

With this Solution Accelerator, you can perform product matching with product metadata and images. Identify duplicates and matches between an initial set of product data and an incremental set of "newly arrived" product data and convert thumbnail images to embeddings leveraging a general-purpose image model from Hugging Face.

Download notebook