Back to results
US 12,749,135 B1Granted
Impact analysis of property
AssigneeUnited Services Automobile Association (USAA)
InventorsRavi Durairaj, Rachel Michelle Ballew, Natalie Kay Helsen, Eric David Schroeder, Amber Nicole Sodikov, David Joaquin Harris
Life of this patent
Estimating the term…
Abstract
A central management system includes one or more processors configured to receive mapping data associated with a target property, identify, via machine learning, a geographic characteristic of interest specific to the target property, determine a modification of the geographic characteristic based on the mapping data, identify an impact of the modification to the target property, and provide an indication of the impact via a graphical user interface.
Claims18
A central management system comprising one or more processors configured to:
- receive mapping data associated with a target property;
- train a machine learning model by submitting to the machine learning model a plurality of training images comprising feature vectors, resulting in a similarity function that predicts fence lines based on examples from the feature vectors;
- identify, via the machine learning model, a geographic characteristic of interest specific to the target property, wherein the geographic characteristic of interest comprises a fence line;
- determine a modification of the geographic characteristic based on the mapping data, wherein the modification to the target property comprises an incorrect demarcation of a legal boundary of the target property based on previously determined locations of the fence line identified via the machine learning model;
- identify an impact of the modification to the target property, wherein identifying the impact comprises:
- retrieve rules from a local database;
- determine that an owner may claim ownership of an additional portion of land within the fence line, wherein the additional portion of land is beyond the legal boundary of the target property;
- and
- automatically provide a notification to the owner via a graphical user interface that the owner may claim ownership of the additional portion of land;
- and
- provide an indication of the impact via the graphical user interface.
8 more elements
A method, comprising:
- receiving mapping data associated with a target property;
- training a machine learning model by submitting to the machine learning model a plurality of training images comprising feature vectors, resulting in a similarity function that predicts fence lines based on examples from the feature vectors;
- identifying, via the machine learning model, a geographic characteristic of interest specific to the target property, wherein the geographic characteristic of interest comprises a fence line;
- determining a modification of the geographic characteristic based on the mapping data, wherein the modification to the target property comprises an incorrect demarcation of a legal boundary of the target property based on previously determined locations of the fence line identified via the machine learning model;
- identifying an impact of the modification to the target property, wherein identifying the impact comprises:
- retrieving rules from a local database;
- determining that an owner may claim ownership of an additional portion of land within the fence line, wherein the additional portion of land is beyond the legal boundary of the target property;
- and
- automatically providing a notification to the owner via a graphical user interface that the owner may claim ownership of the additional portion of land;
- and
- providing an indication of the impact via the graphical user interface.
8 more elements
A non-transitory computer-readable medium comprising computer readable instructions, that when executed by one or more processors, causes the one or more processors to perform operations comprising:
- receiving mapping data associated with a target property;
- training a machine learning model by submitting to the machine learning model a plurality of training images comprising feature vectors, resulting in a similarity function that predicts fence lines based on examples from the feature vectors;
- identifying, via the machine learning model, a geographic characteristic of interest specific to the target property, wherein the geographic characteristic of interest comprises a fence line;
- determining a modification of the geographic characteristic based on the mapping data, wherein the modification to the target property comprises an incorrect demarcation of a legal boundary of the target property based on previously determined locations of the fence line identified via the machine learning model;
- identifying an impact of the modification to the target property, wherein identifying the impact comprises:
- retrieving rules from a local database;
- determining that an owner may claim ownership of an additional portion of land within the fence line, wherein the additional portion of land is beyond the legal boundary of the target property;
- and
- automatically providing a notification to the owner via a graphical user interface that the owner may claim ownership of the additional portion of land;
- and
- providing an indication of the impact via the graphical user interface.
8 more elements
Dependencies are inferred from the claim text.