US 10713672

Here is a concise summary of US patent 10713672:

US Patent 10713672: Discovering neighborhood clusters and uses therefor

• Title: Discovering neighborhood clusters and uses therefor
• Assignee: Carnegie Mellon University
• Inventors: Justin Cranshaw, Raz Schwartz, Jason I. Hong, Norman Sadeh-Koniecpol
• Filing Date: December 18, 2017
• Issue Date: July 14, 2020
• Abstract: The patent describes computer-based systems and methods for identifying neighborhood clusters in a geographic area. These clusters are defined by a mix of venues and are determined using venue check-in data. The mix of venues can be based on social similarity between venues, characteristic neighborhood typologies, temporal check-in patterns, or combinations of these factors. The discovered clusters can then be utilized for various commercial and civic applications.

Plain-Language Overview of Independent Claims:

• Independent Claim 1 (Computer-Implemented Method for Venue Clusters): This claim describes a computer-implemented method for finding two or more groups (clusters) of venues in a geographic area. The method involves:

  1. Collecting "check-in" data from many people at various venues in the area.
  2. For each venue, creating a "check-in intensity vector" that shows how often individual visitors or groups of visitors check into that venue over a specific time.
  3. Creating a "similarity matrix" for all pairs of venues. The similarity score for each pair is calculated based on both how close the venues are geographically and how "socially similar" they are. Social similarity is determined by whether the same people tend to visit both venues.
  4. Using this similarity matrix to identify and group the venues into distinct geographic clusters, where each cluster contains a unique mix of one or more venues.

• Independent Claim 10 (Computer-Implemented Method for Sub-Region Clusters): This claim describes a computer-implemented method similar to Claim 1, but instead of clustering individual venues, it focuses on clustering larger geographic "sub-regions" (like census tracts). The method involves:

  1. Collecting "check-in" data from people at venues within the geographic region, where each venue is assigned to a sub-region.
  2. For each sub-region, creating a "check-in intensity vector" that shows how often individual visitors check into venues within that sub-region over a specific time.
  3. Creating a "similarity matrix" for all pairs of sub-regions. The similarity score for each pair is calculated based on how similar their respective check-in intensity vectors are.
  4. Using this sub-region similarity matrix to identify and group these sub-regions into distinct geographic clusters, where each cluster contains a mix of one or more sub-regions.

• Independent Claim 17 (Computer System for Venue Clusters): This claim describes a physical computer system designed to discover two or more geographic clusters of venues. The system comprises:

  1. A computer database system specifically set up to store venue check-in data from many visitors for many venues.
  2. One or more processors connected to this database. These processors are specifically programmed to perform the same steps outlined in Independent Claim 1:
     • Generate check-in intensity vectors for each venue based on visitor check-in intensity.
     • Generate a pairwise venue similarity matrix, where similarity scores combine geographical distance and social distance (based on common visitors).
     • Identify the geographic clusters of venues using this similarity matrix.

USPTO and CAFC Docket Search Status:

A search of USPTO and CAFC 2026 dockets for patent number 10713672 did not yield specific active litigation or examination details for 2026 in the provided search results. Therefore, I cannot authoritatively confirm any ongoing litigation or other specific docket activity for this patent in 2026 based on the provided search.