Abstract: A method of training and using a machine learning model that controls for consideration of undesired factors which might otherwise be considered by the trained model during its subsequent analyses of new data. For example, the model may be a neural network trained on a set of training images to evaluate an insurance applicant based upon an image or audio data of the insurance applicant as part of an underwriting process to determine an appropriate life or health insurance premium. The model is trained to probabilistically correlate an aspect of the applicant's appearance with a personal and/or health-related characteristic. Any undesired factors, such as age, sex, ethnicity, and/or race, are identified for exclusion. The trained model receives the image (e.g., a “selfie”) of the insurance applicant, analyzes the image without considering the identified undesired factors, and suggests the appropriate insurance premium based only on the remaining desired factors.

Abstract: The present disclosure is directed to apparatuses, systems and methods for automatically classifying images of occupants inside a vehicle. More particularly, the present disclosure is directed to apparatuses, systems and methods for automatically classifying images of occupants inside a vehicle by comparing current image feature data to previously classified image features.

Abstract: A computer-implemented method for improving compression of predictive models can include providing a labeled data set. The computer-implemented method also can include training, using the labeled data set, a neural network model associated with one or more training parameters to create a trained neural network model. Other embodiments are disclosed.

Abstract: A computer-implemented method for providing recommendations includes identifying a set of locations that are of interest to a user and that are within a geographic area. The computer-implemented method also includes identifying one or more businesses based on the set of locations that are of interest to the user, each one of the one or more businesses being within a threshold geographic range of the geographic area. The computer-implemented method further includes transmitting, to an electronic device of the user, indications of the one or more businesses based on the set of locations. Other embodiments are disclosed.

Abstract: A computer-implemented method including receiving an image for evaluation, the image includes video and/or still image data; analyzing the image to extract profile features using a pattern recognition or a facial recognition algorithm, an object recognition, and/or an object character recognition, the profile features include first profile features related to a vehicle and second profile features related to an image verification indicator, the second profile features include one or more features related to an identity of a user; generating an image verification score including a first comparison and a second comparison of at least a portion of one or more of the first profile features and at least a portion of one or more of the second profile features, respectively, and the image verification indicator; and updating a vehicle profile when the image verification score satisfies a threshold of similarity based on the comparisons.

Abstract: Systems and methods for using collecting and analyzing device sensor data to determine whether an individual is an operator or a passenger of a vehicle are disclosed. According to certain aspects, an electronic device associated with the individual may collect or access sensor data that is indicative of or associated with an operation of the vehicle. The electronic device may transmit pertinent portion(s) of the sensor data to a backend server, which may input the portion(s) into a neural network for analysis. The neural network may output a probability metric(s) indicative of whether the individual is a passenger or an operator of the vehicle.

Abstract: A method and system is provided for providing local business-based recommendations or search results to visitors. In response to a request for a digital map of a geographic area, the system determines whether a user is familiar with the geographic area based on the user's location history. In response to determining that the user is unfamiliar with the geographic area and therefore is a visitor of the geographic area, the system provides recommendations, suggestions, or search results to the user which includes local businesses which are of interest to the user. The local businesses may include local businesses which are related to a geographic search query provided by the user, or may include local businesses recommended by the system according to the time of day, time of year, events within the geographic area such as events stored in the user's virtual calendar, reviews of the local businesses, etc.

Abstract: A computer-implemented method for improving compression of predictive models includes generating an unlabeled simulated data set by expanding an initial data set, and generating a labeled data set by predicting the unlabeled, simulated data set using a complex model to output a plurality of labels. The method also includes training a relatively simple neural network using the labeled data set.

Abstract: A computer-implemented method implemented by one or more processors for assessing accident events, the computer-implemented method including: determining, based on first sensor data captured by one or more first sensors associated with a first vehicle, that an accident event occurred; and after determining that the accident event occurred: sending, to an electronic device associated with a second vehicle, a request for second sensor data captured by one or more second sensors of the electronic device associated with the second vehicle; receiving, from the electronic device, the second sensor data; analyzing the second sensor data to determine that the second sensor data is related to the accident event; and aggregating the first sensor data with the second sensor data. Other embodiments are described.

Abstract: A video analytics based image verification system for obtaining initial vehicle profiles is presented. The system may include an external processing server that receives a location of a vehicle and proximate traffic information to determine whether it is safe for a user to obtain an initial vehicle profile. The external processing server may further determine first and second profile features from video data indicative of the vehicle. The external processing server may compare the second profile feature to an image verification indicator to generate an image verification score. A provider server may receive the first profile feature and the image verification score from the external processing server, and update a risk evaluation to include the initial vehicle profile if the image verification score is above an image verification threshold.

Abstract: A computer-implemented method for health assessment includes providing an artificial neural network trained with training data. The method also includes receiving activity data comprising personal activity metrics corresponding to a target person, and determining, by the artificial neural network, a health indicator of the target person based upon the activity data corresponding to the target person. The method further includes outputting the health indicator of the target person to an electronic device for communication of the health indicator to a user of the electronic device. Other embodiments are disclosed.

Abstract: Systems and methods for analyzing documentation for assessing potential fraudulent user submissions are provided. According to certain aspects, a server computer may receive an initial set of documentation descriptive of damage to a property asset, and may analyze the initial set of documentation to determine whether additional documentation is needed. The server computer may initiate a communication channel with a user device via which the additional documentation may be submitted, and the server computer may similarly analyze the additional documentation to determine a likelihood of fraud. The server computer may process the user submission accordingly.

Abstract: Systems and methods for improving vehicular safety are disclosed. According to embodiments, an electronic device may collect or accumulate various sensor data associated with operation of a vehicle by an individual, including image data, telematics data, and/or data indicative of a condition of the individual. The electronic device may analyze the sensor data to determine whether the individual is distracted and whether the vehicle is approaching a location that may be prone to incidents. The electronic device may accordingly generate and present a notification to the individual to mitigate any posed risks.

Abstract: A method of training and using a machine learning model that controls for consideration of undesired factors which might otherwise be considered by the trained model during its subsequent analysis of new data. For example, the model may be a neural network trained on a set of training images to evaluate an insurance applicant based upon an image or audio data of the insurance applicant as part of an underwriting process to determine an appropriate life or health insurance premium. The model is trained to probabilistically correlate an aspect of the applicant's appearance with a personal and/or health-related characteristic. Any undesired factors, such as age, sex, ethnicity, and/or race, are identified for exclusion. The trained model receives the image (e.g., a “selfie”) of the insurance applicant, analyzes the image without considering the identified undesired factors, and suggests the appropriate insurance premium based only on the remaining desired factors.

Abstract: A gate-based vehicle image capture system for analyzing vehicle image data is presented. The system may include a portable imaging gate apparatus configured to capture vehicle image data of a vehicle. The portable imaging gate apparatus may include a plurality of imaging assemblies positioned at a plurality of viewing angles. The system may further include an external processing server configured to receive the vehicle image data from the portable imaging gate apparatus. The external processing server may also analyze the vehicle image data to identify a plurality of vehicle features, and determine a first vehicle feature from the plurality of vehicle features. The first vehicle feature may be related to a vehicle incident. The system may also include a provider server configured to receive the first vehicle feature from the external processing server, and update an aspect of a risk evaluation based on the first vehicle feature.

Abstract: Systems and methods for assessing accident events are disclosed According to certain aspects, an electronic device may access sensor data associated with operation of a vehicle, and analyze the sensor data to determine that an accident event has occurred. In response, the electronic device may dynamically and automatically retrieve additional sensor data from at least one additional electronic device via a short range communication. The electronic device may compile or aggregate the additional sensor data with the original sensor data to provide a thorough representation of the accident event.

Abstract: Systems and methods for improving vehicular safety are disclosed. According to embodiments, an electronic device may collect or accumulate various sensor data associated with operation of a vehicle by an individual, including image data, telematics data, and/or data indicative of a condition of the individual. The electronic device may analyze the sensor data to determine whether the individual is distracted and whether the vehicle is approaching a location that may be prone to incidents. The electronic device may accordingly generate and present a notification to the individual to mitigate any posed risks.

Abstract: A system comprising one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform operations: comparing, by an electronic device, at least one parameter of a set of telematics data associated with a vehicle and at least one additional parameter of an additional set of telematics data associated with one or more other vehicles within a threshold distance of the vehicle to determine a difference between the at least one parameter and the at least one additional parameter; and displaying, by the electronic device, the difference to the operator of the vehicle to notify the operator of the vehicle about the difference. Other embodiments are described.

Abstract: Apparatuses, systems and methods are provided for determining vehicle driver distractions. More particularly, apparatuses, systems and methods are provided for receiving vehicle interior data from at least one vehicle interior sensor, wherein the vehicle interior data is representative of at least one distraction of at least one vehicle occupant that is associated with a position of a body of the at least one vehicle occupant; generating vehicle driving route distraction data based on the vehicle interior data; and providing instructions to the at least one vehicle occupant to return to a normal driving position.

Abstract: A method of training and using a machine learning model that controls for consideration of undesired factors which might otherwise be considered by the trained model during its subsequent analysis of new data. For example, the model may be a neural network trained on a set of training images to evaluate an insurance applicant based upon an image or audio data of the insurance applicant as part of an underwriting process to determine an appropriate life or health insurance premium. The model is trained to probabilistically correlate an aspect of the applicant's appearance with a personal and/or health-related characteristic. Any undesired factors, such as age, sex, ethnicity, and/or race, are identified for exclusion. The trained model receives the image (e.g., a “selfie”) of the insurance applicant, analyzes the image without considering the identified undesired factors, and suggests the appropriate insurance premium based only on the remaining desired factors.