Allstate Insurance Company

Meg G Walters

Dana Ferguson

Eric D Huls

Grady Irey

Caryl M Styrsky

Christine L Bischoff

William Polisson

Ralph Adam Benjamin Tyner

Charles F Cox

Jean Utke

Yusuf Mansour

Tia Miceli

Dylan Stark

Adam Sturt

Jeremy Werner

Adam Austin

Amber Parsons

Xiaolan Wu

Ryan Rosenberg

Lizette Lemus Gonzalez

Weizhou Wang

Stephanie Wong

Lakshmi Prabha Nattamai Sekar

Emily Pavey

Deborah-Anna Reznek

Mark V Slusar

Mark S Richards

Nilesh Malpekar

Anna Reifman

Gabriel Federico Carballo Nunez

Methods, computer-readable media, software, and apparatuses may assist in assessing proportional fault in an automobile accident involving an automobile having one or more autonomous features. An expected behavior of an autonomous feature is compared to an observed outcome of an accident and a fault proportion between a human driver and the autonomous feature may be determined, based on the comparison.

Assessing proportional fault in an automobile accident

Systems and methods for deployment of a protective component, generation of a customized design for the protective component, or combinations thereof are associated with a structure comprising a portion, a neural network model, processor(s), and memory storing machine readable instructions. When executed for deployment, the neural network model predicts the likelihood of the occurrence of the natural event in the geographic area within the time frame as high as defined by when the likelihood is above a threshold, and deploys the protective component for protecting the portion of the structure when the likelihood is high. For customized design, the neural network model is used to access dimension and weather data associated with a structure and weather data to generate the customized design of the protective component for the structure.

Intelligent structural protection systems and methods

Aspects of the disclosure relate to using computer vision methods for asset evaluation. A computing platform may receive historical images of a plurality of properties and corresponding historical inspection results. Using the historical images and historical inspection results, the computing platform may train a roof waiver model (which may be a computer vision model) to output inspection prediction information directly from an image. The computing platform may receive a new image corresponding to a particular residential property. Using the roof waiver model, the computing platform may analyze the new image to output of a likelihood of passing inspection. The computing platform may send, to a user device and based on the likelihood of passing inspection, inspection information indicating whether or not a physical inspection should be performed and directing the user device to display the inspection information, which may cause the user device to display the inspection information.

Computer vision methods for loss prediction and asset evaluation based on aerial images

Aspects of the disclosure relate to using computer vision methods to forecast damage. A computing platform may receive historical images comprising aerial images of residential properties and historical loss data corresponding to the residential properties. Using the historical images and the historical loss data, the computing platform may train a computer vision model, which may configure the computer vision model to output loss prediction information directly from an image. The computing platform may receive a new image corresponding to a particular residential property, and may analyze the new image, using the computer vision model, which may directly result in a likelihood of damage score. Based on the likelihood of damage score, the computing platform may send likelihood of damage information and one or more commands directing a user device to display the likelihood of damage information, which may cause the user device to display the likelihood of damage information.

Methods, computer-readable media, systems and apparatuses for determining and implementing dynamic usage-based insurance policies are presented. A cost per day and a cost per mile associated with the dynamic usage-based insurance policy may be determined. The cost per mile may be computed independently for each of a plurality of road segments comprising a trip based on a time of day that the vehicle traveled each road segment, a road type associated with each road segment. The cost per mile of each road segment of the driving trip may further include whether or not hard braking or hard cornering events were encountered during a specified cumulative driving distance that may include, at least a portion, of a plurality of driving trips. The cost per mile for each of the plurality of driving trips may be adjusted once a threshold distance has been traveled by the vehicle based on the number of hard braking events and hard cornering events occurred during that distance.

Dynamic usage-based policies

Growing community of inventors

Evanston, IL, United States of America

Meg G Walters