Patent Application for Linking Physical Locations and Online Channels in a Database

Strategic Insights -

A patent application filed on June 18, 2025, by inventors FENG, Fan, FENICHEL, Allison, and REED, Illiana, proposes a system for linking physical location data and online channel data of entities in a database. The system utilizes natural language processing, machine learning, and graph theory to identify and associate online locations with physical locations and brands.

The inventors' proposed system includes a device that performs natural language processing to extract features from unstructured data, iteratively trains a machine learning model to generate a trained model, and applies the trained model to determine the likelihood of an online location being associated with a brand. The system also generates credentials for completing transactions at online or physical locations associated with the brand.

The patent application claims a device, method, and non-transitory computer-readable medium for implementing the system. The claims describe various aspects of the system, including the use of machine learning models, such as neural networks or support vector machines, and the generation of temporary credentials.

The inventors' summary highlights the importance of linking physical and online locations in a database, stating that "In some implementations, a system for linking physical location data and online channel data of entities in a database includes one or more memories; and one or more processors, communicatively coupled to the one or more memories, configured to: receive information associated with a plurality of locations, where the plurality of locations include at least one in-person location and at least one online location; identify, from the plurality of locations, candidate locations, parse information associated with the candidate locations to identify information for a set of features, where the set of features include at least one of a name, a phone number, a geographic location, a uniform resource locator, a headquarter address, an entity category, metadata, or transaction data; apply a machine learning model to the information for the set of features for the candidate locations; determine whether the candidate locations are associated with a same entity based on an output of the machine learning model; and add the candidate locations to a graph associated with the entity if the candidate locations are associated with the entity, where the graph associated with the entity indicates at least one online location associated with the entity and at least one in-person location associated with the entity."

Key Takeaways:

  • The patent application proposes a system for linking physical location data and online channel data of entities in a database using natural language processing, machine learning, and graph theory.
  • The system utilizes a device that performs natural language processing to extract features from unstructured data, iteratively trains a machine learning model, and applies the trained model to determine the likelihood of an online location being associated with a brand.
  • The system generates credentials for completing transactions at online or physical locations associated with the brand.
  • The patent application claims a device, method, and non-transitory computer-readable medium for implementing the system.
  • The claims describe various aspects of the system, including the use of machine learning models, such as neural networks or support vector machines, and the generation of temporary credentials.
  • The inventors' summary highlights the importance of linking physical and online locations in a database.
  • The system can be used in various industries, including e-commerce, retail, and finance.

Statistics:

  • The patent application was filed on June 18, 2025.
  • The patent application was posted online on October 9, 2025.
  • The system proposed in the patent application utilizes natural language processing, machine learning, and graph theory.
  • The system can generate credentials for completing transactions at online or physical locations associated with a brand.
  • The system can be implemented using a device, method, and non-transitory computer-readable medium.

Sources:

  • FENG, Fan; FENICHEL, Allison; REED, Illiana. Linking Physical Locations And Online Channels In A Database. U.S. Patent Application Number 20250315848, filed June 18, 2025 and posted October 9, 2025. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(20250315848)&db=US-PGPUB&type=ids