Patent Application for Hierarchical Product Representation from Purchase Orders

ChangeBridge: Patent Apps - Business Methods (G06Q)

Detected March 26th, 2026

Summary

The USPTO has published a patent application (US20260087431A1) detailing systems and methods for generating a hierarchical product representation from purchase orders. The application, filed on December 16, 2024, describes a process involving embeddings, clustering, and labeling to create product hierarchies.

View original document View source feed page

What changed

This document is a published patent application from the USPTO, specifically application US20260087431A1, titled "SYSTEMS AND METHODS FOR GENERATING A HIERARCHICAL REPRESENTATION OF A PLURALITY OF PRODUCTS." The application, filed on December 16, 2024, by inventors Simon Dobers et al., outlines a method for generating a hierarchical product representation by receiving purchase orders, creating embeddings based on semantic similarity, clustering these embeddings, and assigning labels to form tiers in the hierarchy. Subsequent iterations can build additional tiers.

As this is a patent application, it does not impose immediate regulatory obligations or compliance deadlines on businesses. However, companies involved in e-commerce, retail, or data analytics may find the described methods relevant for internal product categorization, inventory management, or market analysis. The application is currently in a draft stage, and its eventual granting could impact intellectual property landscapes in business methods related to product representation and data processing.

Source document (simplified)

← USPTO Patent Applications

SYSTEMS AND METHODS FOR GENERATING A HIERARCHICAL REPRESENTATION OF A PLURALITY OF PRODUCTS

Application US20260087431A1 Kind: A1 Mar 26, 2026

Inventors

Simon Dobers, Sebastian Münchow, Andy Chen, Leo Razumovsky, Daisy Sellem, Jonas Bernsdorff

Abstract

In some examples, systems and methods for generating a hierarchical representation are provided. An example method includes receiving a plurality of purchase orders, generating a plurality of embeddings based on the plurality of purchase orders, clustering the plurality of embeddings into one or more embedding subsets based on a semantic similarity between each embedding of the plurality of embeddings and each other embedding of the plurality of embeddings, and identifying a plurality of labels for the plurality of embedding subsets. In some examples, the method includes generating the hierarchical representation such that the plurality of labels forms a tier in the hierarchical representation. In some examples, subsequent iterations of one or more aspects of the method can be performed to generate additional tiers in the hierarchical representation, where the plurality of labels from a preceding iteration can be used to generate the embeddings for a current iteration.