Prior art

To identify the most relevant prior art for US patent 10713672, I will examine the patent citations listed on the patent document itself. The patent document for US10713672B1 lists several "Prior art documents" under the "Cited by" section on Google Patents. These include both patent documents and non-patent literature.

Here's an analysis of some of the most relevant prior art documents cited in US10713672:

1. U.S. Patent Application Publication No. 2011/0270634 A1 (Cranshaw et al.)

• Full Citation: US 2011/0270634 A1
• Publication/Filing Date: Publication: November 3, 2011. Filing: April 29, 2011.
• Brief Description: This patent application describes systems and methods for discovering activity patterns and common paths taken by users of a location-based social network. It focuses on identifying frequently traveled routes and associated activities based on check-in data. This could be relevant to understanding user movement and density in geographic areas.
• Potential Anticipation (35 U.S.C. § 102): This reference could potentially anticipate aspects of claims related to the collection and use of venue check-in data and the understanding of user movement patterns. Specifically, the concept of utilizing location-based social network data for analyzing user behavior in geographic areas, which forms a basis for the clustering in US10713672, might be challenged. Given that US10713672 explicitly mentions "venue check-in apps" and the collection of data from such sources, US2011/0270634 A1's focus on location-based social network data is highly pertinent. It could potentially anticipate the "collecting venue check-in data" step of claims 1 and 10, and the system for storing such data in claim 17.

2. U.S. Patent No. 8,364,577 B2 (Schwartz et al.)

• Full Citation: US 8,364,577 B2
• Publication/Filing Date: Issue: January 29, 2013. Filing: February 14, 2011.
• Brief Description: This patent describes methods and systems for recommending venues to users based on user preferences and location data. It might involve analyzing user check-in history and the characteristics of venues to make recommendations.
• Potential Anticipation (35 U.S.C. § 102): While US8,364,577 B2 focuses on recommendations, the underlying techniques for collecting and analyzing venue and user data, including check-in data and user preferences, could be relevant. The patent US10713672 mentions "venue rating system" and "venue review system" as sources of check-in data, which aligns with the data types that would be used for recommendation systems. This could potentially anticipate aspects of the data collection and initial processing described in claims 1, 10, and 17, particularly where user preferences and interactions with venues are considered.

3. U.S. Patent Application Publication No. 2012/0158498 A1 (Cranshaw et al.)

• Full Citation: US 2012/0158498 A1
• Publication/Filing Date: Publication: June 21, 2012. Filing: December 16, 2011.
• Brief Description: This publication relates to systems and methods for automatically detecting and characterizing social events using location-based social network data. This involves identifying gatherings of users at specific locations and times.
• Potential Anticipation (35 U.S.C. § 102): The detection and characterization of social events based on location-based social network data in US2012/0158498 A1 directly involves the analysis of user presence at venues over time. This is highly relevant to the "check-in intensity vector" and "temporal check-in pattern types" described in US10713672. It could potentially anticipate the generation of check-in intensity vectors and the use of temporal data for clustering, as mentioned in claims 1 and 10. The concept of "social similarity" based on common users visiting venues (as described in claim 1) could also find some basis in the techniques for identifying social events.

Non-Patent Literature Examples (if available, a comprehensive list would require direct access to the full USPTO file wrapper):

The patent description itself mentions "Cheng et al. (“Exploring millions of footprints in location sharing services,” AAAI ICWSM, 2011)" and "D. M. Blei and P. I. Frazier, “Distance dependent Chinese restaurant processes,” J. Mach. Learn. Res., 2461-2488, November 2011" and "Ghosh et al., “Spatial distance dependent Chinese restaurant processes for image segmentation,” Neural Information Processing Systems, 2011". These are also critical pieces of prior art.

• Cheng et al. (“Exploring millions of footprints in location sharing services,” AAAI ICWSM, 2011)

  • Publication Date: 2011
  • Brief Description: This paper discusses exploring millions of footprints in location-sharing services, which directly pertains to the large-scale analysis of user check-in data from social networks. The inventors of US10713672 explicitly state they used data released by Cheng et al.
  • Potential Anticipation (35 U.S.C. § 102): This work likely demonstrates the prior art for collecting, preprocessing, and generally utilizing large datasets of location-based social network check-ins for analysis. This directly impacts the novelty of the "collecting venue check-in data" step in claims 1 and 10, and the associated data storage system in claim 17. The general concept of deriving insights from such "footprints" would be known prior art.

• D. M. Blei and P. I. Frazier, “Distance dependent Chinese restaurant processes,” J. Mach. Learn. Res., 2461-2488, November 2011

  • Publication Date: November 2011
  • Brief Description: This paper introduces the Distance Dependent Chinese Restaurant Process (ddCRP), a non-parametric Bayesian method for clustering non-exchangeable data, explicitly used in US10713672 for its clustering methodology.
  • Potential Anticipation (35 U.S.C. § 102): The core clustering methodology for "neighborhood typologies" in US10713672 relies heavily on the ddCRP and its hierarchical extensions. Claims that involve determining the mix of venues "based on patterns of venue category type in the venue category data emblematic of a neighborhood type" and using "inference to compute a probabilistic distribution of venues for each cluster" or "statistical sampling, such as Gibbs sampling" (as seen in dependent claims of claim 1, and also relevant for claim 10) are directly informed by this prior art. The novelty of using ddCRP for general clustering would be anticipated by this work.

• Ghosh et al., “Spatial distance dependent Chinese restaurant processes for image segmentation,” Neural Information Processing Systems, 2011

  • Publication Date: 2011
  • Brief Description: This paper extends the ddCRP to hierarchical modeling, specifically for image segmentation, and is also explicitly cited by the inventors of US10713672 as a basis for their Gibbs sampler.
  • Potential Anticipation (35 U.S.C. § 102): Similar to the Blei and Frazier paper, this work directly anticipates aspects of the probabilistic modeling and inference (e.g., Gibbs sampling) used in US10713672, especially for the hierarchical application of ddCRP. Claims related to probabilistic models and statistical sampling for determining clusters, particularly those involving hierarchical structures or spatial considerations, would be impacted.