Prior art

To identify the most relevant prior art for US patent 11935082, I need to access the patent's full citation details and its list of cited references from the USPTO database. Since I cannot directly interact with a live USPTO search portal, I will rely on the provided patent text to extract the cited prior art references.

Based on the full patent text provided, I will extract information on cited prior art. The "Prior art keywords" section on Google Patents often lists keywords from prior art, but for specific patent citations, I need to look at the "References Cited" section of the full patent document. The provided text is a detailed description of the patent, but it doesn't explicitly list a "References Cited" section with full patent citations in a structured manner typical of a USPTO patent document. However, it does mention prior art and related research papers in the description.

Specifically, the patent description mentions the following prior art in relation to its methods:

1. "Distance dependent Chinese restaurant processes," by D. M. Blei and P. I. Frazier, J. Mach. Learn. Res., 2461-2488, November 2011. This paper is explicitly incorporated by reference and is foundational to the ddCRP modeling used in the patent for discovering neighborhood typologies.

   • Publication Date: November 2011
   • Description: This paper introduces the distance dependent Chinese restaurant process (ddCRP), which specifies a distribution over partitions of non-exchangeable data, used as a nonparametric prior over mixture components.
   • Potential Anticipation: This reference potentially anticipates claims related to the use of distance-dependent Chinese restaurant processes for clustering, particularly claims involving probabilistic models for determining neighborhood typologies based on venue category data or temporal check-in patterns. For instance, the general concept of using ddCRP for clustering and defining partitions of data could be considered.

2. "Spatial distance dependent Chinese restaurant processes for image segmentation," by Ghosh et al., Neural Information Processing Systems, 2011. This work extends the ddCRP to hierarchical modeling and is also incorporated by reference.

   • Publication Date: 2011
   • Description: This paper extends the ddCRP to hierarchical modeling, where observations in different groups are linked by sharing neighborhood parameters across cities.
   • Potential Anticipation: This reference could potentially anticipate claims relating to the hierarchical application of ddCRP, especially for segmenting data based on spatial distances, which the patent applies to clustering venues or sub-regions. Claims involving the use of hierarchical probabilistic models for determining neighborhood clusters could be affected.

3. "Exploring millions of footprints in location sharing services," by Cheng et al., AAAI ICWSM, 2011. This paper is cited as the source for approximately 11 million Foursquare check-ins used in the inventors' experiments.

   • Publication Date: 2011
   • Description: This publication describes the extraction of check-in data from various location-based social networks, specifically Twitter public feeds, which was used in the experimental validation of the claimed invention.
   • Potential Anticipation: While not directly describing a clustering methodology, this reference demonstrates the availability and use of large-scale venue check-in data from social media platforms prior to the patent's priority date. It could be relevant to the data collection and input aspects of the claims, specifically regarding the "storing venue check-in data from multiple venue visitors for multiple venues in the geographic region" element.

4. "From deformations to parts: Motion-based segmentation of 3d objects," by Ghosh et al., Advances in Neural Information Processing Systems 25, pp. 2006-2014, 2012. This reference is mentioned as providing portions of the ddCRP Gibbs sampler used in the MATLAB implementation for the patent's experiments.

   • Publication Date: 2012
   • Description: This paper describes a ddCRP Gibbs sampler for 3D Mesh segmentation, which was adapted and extended for the hierarchical model used in the patent's experiments.
   • Potential Anticipation: This reference could be relevant to the specific implementation details of the Gibbs sampling method for clustering, particularly if any claims were to focus on the particular mechanics of the sampler itself. However, the patent emphasizes its adaptation of this work for a different purpose (neighborhood clustering).

It's important to note that the patent description highlights how its methods build upon and extend these prior art techniques, particularly the ddCRP and its hierarchical extensions, by applying them to the specific problem of discovering neighborhood clusters based on venue check-in data in a novel way that considers geographical proximity, social similarity, neighborhood typologies, and temporal patterns. Therefore, while these references describe foundational techniques, the patent claims aim to distinguish the invention by its specific application and combinations of these techniques for neighborhood clustering.