Hail severity predictions using artificial intelligence
The disclosure includes systems and methods for identifying, using one or more processors, a set of properties that experienced hail damage; determining, using the one or more processors, a first set of features associated with each property in the set of properties, the first set of features including one or more hail features at a location of the property, one or more property features describing an associated property, and one or more damage values; training, using the one or more processors, a first damage severity model; and validating, using the one or more processors, the first damage severity model.
1 . A computer implemented method comprising:
determining, using one or more processors, a set of training data for training a first damage severity model including:
identifying a set of properties that experienced hail damage, the set of training data for the first damage severity model including the identified set of properties that experienced hail damage;
determining a first set of features associated with each property in the set of properties, the first set of features including one or more hail features at a location of the property, one or more property features describing an associated property, and one or more damage values;
determining a gain associated with each feature in the first set of features; and
determining a first subset of features from the first set of features by eliminating a second subset of features within the first set of features from the first set of training data for the first damage severity model, based on a gain threshold, wherein the first subset of features are associated with sufficient gain values to satisfy the gain threshold and the second subset of features are associated with insufficient gain values to satisfy the gain threshold;
training, using the one or more processors, the first damage severity model based on the set of training data for the first damage severity model; and
validating, using the one or more processors, the first damage severity model.
2 . The computer implemented method of claim 1 further comprising:
receiving a location;
applying the first damage severity model; and
presenting a damage severity.
3 . The computer implemented method of claim 2 , wherein the damage severity is one or more of: a roof area damaged, a claim cost, and a cost of replacement.
4 . The computer implemented method of claim 2 further comprising:
determining, based on a determined damage severity, one or more of: a remedial action to reduce a risk of hail; whether to approve or deny hail insurance coverage or an insurance claim; an insurance premium associated with the location, an adjustment to an insurance premium associated with location; and to warn one or more of a property owner, resident, financier and insurer associated with the location of a risk of damage posed by hail.
5 . The computer implemented method of claim 1 , wherein the set of properties that experienced hail damage includes one or more of a first property from hail claims data and a second property from building permit data, where a building permit associated with the second property indicates hail.
6 . The computer implemented method of claim 1 , wherein identifying the set of properties that experienced hail damage comprises:
obtaining an initial set of properties that experienced hail damage;
eliminating, from the initial set of properties, properties unassociated with a damage value;
eliminating, from the initial set of properties, one or more properties under construction; and
identifying, as the set of properties that experienced hail damage, a remaining set of properties.
7 . The computer implemented method of claim 6 , wherein a threshold is associated with one or more of a maximum roof size, a minimum roof size, a maximum damage value, a minimum damage value, and an outlier.
8 . The computer implemented method of claim 1 , wherein the first damage severity model is a regression model.
9 . The computer implemented method of claim 1 , wherein the first damage severity model is a classifier.
10 . The computer implemented method of claim 1 , wherein the first set of features includes one or more of a roof area, a building area, a vegetation density, a roof material, a roof quality, a roof height, a roof pitch, presence of a skylight, a portion of a roof covered by skylights, presence of a solar panel, a portion of the roof covered by solar panels, a number of roof facets, a roof shape, a land cover code, a temperature, a precipitation type or metric, and an elevation.
11 . A system comprising:
a processor; and
a memory, the memory storing instructions that, when executed by the processor, cause the system to:
determine a set of training data for training a first damage severity model including:
identify a set of properties that experienced hail damage, the set of training data for the first damage severity model including the identified set of properties that experienced hail damage;
determine a first set of features associated with each property in the set of properties, the first set of features including one or more hail features at a location of the property, one or more property features describing an associated property, and one or more damage values;
determine a gain associated with each feature in the first set of features; and
determine a first subset of features from the first set of features by eliminating a second subset of features within the first set of features from the first set of training data for the first damage severity model, based on a gain threshold, wherein the first subset of features are associated with sufficient gain values to satisfy the gain threshold and the second subset of features are associated with insufficient gain values to satisfy the gain threshold;
train a first damage severity model based on the set of training data for the first damage severity model; and
validate the first damage severity model.
12 . The system of claim 11 , the memory further storing instructions that, when executed by the processor, cause the system to:
receive a location;
apply the first damage severity model; and
present a damage severity.
13 . The system of claim 12 , wherein the damage severity is one or more of: a roof area damaged, a claim cost, and a cost of replacement.
14 . The system of claim 12 , the memory further storing instructions that, when executed by the processor, cause the system to:
determine, based on a determined damage severity, one or more of: a remedial action to reduce a risk of hail; whether to approve or deny hail insurance coverage or an insurance claim; an insurance premium associated with the location, an adjustment to an insurance premium associated with location; and to warn one or more of a property owner, resident, financier and insurer associated with the location of a risk of damage posed by hail.
15 . The system of claim 11 , wherein the set of properties that experienced hail damage includes one or more of a first property from hail claims data and a second property from building permit data, where a building permit associated with the second property indicates hail.
16 . The system of claim 11 , wherein identifying the set of properties that experienced hail damage comprises:
obtaining an initial set of properties that experienced hail damage;
eliminating, from the initial set of properties, properties unassociated with a damage value;
eliminating, from the initial set of properties, one or more properties under construction; and
identifying, as the set of properties that experienced hail damage, a remaining set of properties.
17 . The system of claim 16 , wherein a threshold is associated with one or more of a maximum roof size, a minimum roof size, a maximum damage value, a minimum damage value, and an outlier.
18 . The system of claim 11 , wherein the first damage severity model is a regression model.
19 . The system of claim 11 , wherein the first damage severity model is a classifier.
20 . The system of claim 11 , wherein the first set of features includes one or more of a roof area, a building area, a vegetation density, a roof material, a roof quality, a roof height, a roof pitch, presence of a skylight, a portion of a roof covered by skylights, presence of a solar panel, a portion of the roof covered by solar panels, a number of roof facets, a roof shape, a land cover code, a temperature, a precipitation type or metric, and an elevation.