Systems and methods for determining building damage

1. A damage assessment (DA) computing device for determining roof damage to a building, the DA computing device comprising a processor and a memory communicatively coupled to the processor, wherein the processor is programmed to:

train a machine learning damage model using historical damage data associated with a plurality of historical weather damage incidents for a respective plurality of buildings, the historical damage data including parameters associated with the respective building, a respective historical weather event, and a known damage status representing an extent of damage to the respective building following the historical weather event;

identify a plurality of buildings that are susceptible to potential damage from an upcoming weather event, the plurality of buildings including the building;

input data associated with a roof of the building to the trained damage model, the data including a plurality of parameters associated with the building and with the upcoming weather event;

receive a model output from the trained damage model, the model output including a damage status of the roof representing the predicted extent of damage to the roof;

when the predicted extent of damage to the roof exceeds a threshold, automatically generate a claim initiation message including instructions that cause display of the model output including the damage status, a first link that, upon selection thereof, causes initiation of an insurance claim for the roof based upon the model output and the parameters associated with the building, and a second link that, upon selection thereof, declines initiation of the insurance claim;

transmit the claim initiation message to a remote computing device of a user associated with the building for display of the damage status, the first link, and the second link on a graphical user interface of the computing device;

store the model output and an identification of whether the first or second link was selected, as updated parameters; and

re-train the damage model using the updated parameters.

2. The DA computing device of claim 1 , wherein the processor is further programmed to:

receive user input from the remote computing device associated with the user, the user input indicating selection of the first link; and

automatically initiate the insurance claim for the roof.

3. The DA computing device of claim 1 , wherein the processor is further programmed to:

receive user input from the remote computing device associated with the user, the user input including (i) text data, image data, video data, or audio data associated with the roof of the building, and (ii) a selection of the link, indicating approval of the user to proceed with the insurance claim based upon the model output and the text data, image data, video data, or audio data; and

automatically initiate the insurance claim for the roof.

4. The DA computing device of claim 3 , wherein, to automatically initiate the insurance claim for the roof, the processor is further programmed to automatically populate an insurance claim form using the user input and the model output.

5. The DA computing device of claim 1 , wherein the processor is further programmed to identify the building for a proactive damage assessment by:

detecting the upcoming weather event in a geographic region including the building; and

performing a lookup in a database using an identifier of the geographic region to identify the plurality of buildings susceptible to potential damage from the upcoming weather event.

6. The DA computing device of claim 1 , wherein the processor is further programmed to:

receive the model output further including a confidence score; and

determine the predicted damage to the roof exceeds the threshold based upon the model output.

7. A computer-based method for determining damage of a roof, said method implemented using a damage assessment (DA) computing device including a processor and a memory communicatively coupled to the processor, said method comprising:

training, by the processor, a machine learning damage model using historical damage data associated with a plurality of historical weather damage incidents for a respective plurality of buildings, the historical damage data including parameters associated with the respective building, a respective historical weather event, and a known damage status representing an extent of damage to the respective building following the historical weather event;

identifying, by the processor, a plurality of buildings that are susceptible to potential damage from an upcoming weather event, the plurality of buildings including the building;

inputting, by the processor, data associated with a roof of the building to the trained damage model, the data including a plurality of parameters associated with the building and with the upcoming weather event;

receiving, by the processor, a model output from the trained damage model, the model output including a damage status of the roof representing the predicted extent of damage to the roof;

when the predicted extent of damage to the roof exceeds a threshold, automatically generating, by the processor, a claim initiation message including instructions that cause display of the model output including the damage status, a first link that, upon selection thereof, causes initiation of an insurance claim for the roof based upon the model output and the parameters associated with the building, and a second link that, upon selection thereof, declines initiation of the insurance claim;

transmitting, by the processor, the claim initiation message to a remote computing device of a user associated with the building for display of the damage status, the first link, and the second link on a graphical user interface of the computing device;

storing the model output and an identification of whether the first or second link was selected, as updated parameters; and

re-training the damage model using the updated parameters.

8. The method of claim 7 , further comprising:

receiving, by the processor, user input from the remote computing device associated with the user, the user input indicating selection of the first link; and

automatically initiating, by the processor, the insurance claim for the roof.

9. The method of claim 7 , further comprising:

receiving, by the processor, user input from the remote computing device associated with the user, the user input including (i) text data, image data, video data, or audio data associated with the roof of the building, and (ii) a selection of the link, indicating approval of the user to proceed with the insurance claim based upon the model output and the text data, image data, video data, or audio data; and

automatically initiating, by the processor, the insurance claim for the roof.

10. The method of claim 9 , wherein automatically initiating the insurance claim comprises automatically populating, by the processor, an insurance claim form using the user input and the model output.

11. The method of claim 7 , wherein the processor identifying the building for a proactive damage assessment comprises:

detecting, by the processor, the upcoming weather event in a geographic region including the building; and

performing, by the processor, a lookup in a database using an identifier of the geographic region to identify the plurality of buildings susceptible to potential damage from the upcoming weather event.

12. The method of claim 7 , further comprising:

receiving, by the processor, the model output further including a confidence score; and

determining, by the processor, the predicted damage to the roof exceeds the threshold based upon the model output.

13. At least one non-transitory computer-readable storage media having computer-executable instructions embodied thereon, wherein when executed by a processor, the computer-executable instructions cause the processor to:

train a machine learning damage model using historical damage data associated with a plurality of historical weather damage incidents for a respective plurality of buildings, the historical damage data including parameters associated with the respective building, a respective historical weather event, and a known damage status representing an extent of damage to the respective building following the historical weather event;

identify a plurality of buildings that are susceptible to potential damage from an upcoming weather event, the plurality of buildings including the building;

input data associated with a roof of the building to the trained damage model, the data including a plurality of parameters associated with the building and with the upcoming weather event;

receive a model output from the trained damage model, the model output including a damage status of the roof representing the predicted extent of damage to the roof;

when the predicted extent of damage to the roof exceeds a threshold, automatically generate a claim initiation message including instructions that cause display of the model output including the damage status, a first link that, upon selection thereof, causes initiation of an insurance claim for the roof based upon the model output and the parameters associated with the building, and a second link that, upon selection thereof, declines initiation of the insurance claim;

transmit the claim initiation message to a remote computing device of a user associated with the building for display of the damage status, the first link, and the second link on a graphical user interface of the computing device;

store the model output and an identification of whether the first or second link was selected, as updated parameters; and

re-train the damage model using the updated parameters.

14. The non-transitory computer-readable storage media of claim 13 , wherein the instructions further cause the processor to:

receive user input from the remote computing device associated with the user, the user input indicating selection of the first link; and

automatically initiate the insurance claim for the roof.

15. The non-transitory computer-readable storage media of claim 13 , wherein the instructions further cause the processor to:

receive user input from the remote computing device associated with the user, the user input including (i) text data, image data, video data, or audio data associated with the roof of the building, and (ii) a selection of the link, indicating approval of the user to proceed with the insurance claim based upon the model output and the text data, image data, video data, or audio data; and

automatically initiate the insurance claim for the roof.

16. The non-transitory computer-readable storage media of claim 15 , wherein the instructions further cause the processor to automatically initiate the insurance claim for the roof by automatically populating an insurance claim form using the user input and the model output.

17. The non-transitory computer-readable storage media of claim 13 , wherein the instructions further cause the processor to:

receive the model output further including a confidence score; and

determine the predicted damage to the roof exceeds the threshold based upon the model output.