Abstract: A system and method for generating virtual financial accounts that may be used to consummate transactions when the mobile device is off-line. The virtual financial accounts are generated in mobile devices that are associated with financial institutions. The financial accounts may be credit card or debit card accounts. Each attempted financial transaction by the mobile device is treated as a unique event. An app running on the mobile device calculates a risk score for the attempted transaction based on factors such as GPS location and typical spending patterns for the financial account, and then approves or denies the transaction. In some cases, biometric verification devices (such as fingerprints, facial recognition or voice recognition) may be used by the app to further authenticate the transaction.

Abstract: Systems and methods for digital integration of financial features between a user and a service provider is provided. For example, if a user device is registered with the service provider, the user device may receive an electronic notification (e.g., calendar entry) related to outstanding account operations. The user device may be able to complete the outstanding account operations by accessing the electronic notification and embedded financial information.

Abstract: Systems and methods for virtual walkthroughs are provided. Pre-loss and post loss captures of an environment are captured and analyzed to identify loss estimates and trigger claim fulfillment based upon the loss estimates.

Abstract: A system and method for training a security model in a testing system using machine learning is disclosed. For example, the method includes accessing information about security attacks from at least one security attack information source, recognizing types of security attacks being attempted using natural language processing to process the information about security attacks, attempting at least one of the recognized types of security attacks on a network, identifying at least one security vulnerability based on results of the attempting, training the security model based on the identified at least one security vulnerability using machine learning, providing the trained security model to a user of the network, and blocking a security attack using the trained security model.

Abstract: A user authorization device (or UAD) physically connected to a user's router to aid in the authentication of the user. A bank server polls the user's router to verify that the UAD is connected to the router before granting the user the capability to perform transactions on the user's account that would be otherwise restricted. The bank server implements a set of policies designed by the user that require the UAD to be present for some transactions with the account to be performed. The presence of this device thus adds an extra layer of security to such financial transactions beyond the use of a username and password.

Abstract: Systems and methods for detecting use of a property are provided. A first set of property identifying data for a first property is received from a first source. The first source indicates a use of the first property but does not publicly provide an address of the first property prior to initiation of a request for use of the first property. A second set of property identifying data for a second property from a second source is also received. The second source provides a property address of the second property. The first set of data are compared to the second set of data, to determine an amount of overlap. When the amount of overlap meets a pre-defined threshold amount, the first property and the second property are identified as a common property and a notice is provided of the property address and the use of the common property.

Abstract: Systems and methods for fault tolerant digital payments are disclosed. The system comprises a computing system that can automatically determine if payments should be processed using a digital payment processing system, or using a blockchain. When the system detects that communication with the digital payment processing system has been interrupted, the system may facilitate transactions through the blockchain. In particular, the system can retrieve private keys for addresses on the blockchain and facilitate customer payments through the blockchain.

Abstract: A method and system for improving and automating the identification of homes located in an imminent or current disaster zone and causing an immediate release of funds to residents without requiring the insured member to first submit a claim are disclosed. The system and method are configured to determine whether there is a high likelihood of a disaster event occurring in a region, based on external sources or local device sensors. In response to the disaster event, a smart contract is executed such that funds are expeditiously released via a distributed ledger system to insured members in the first region.

Abstract: Embodiments are described herein for a teller check that includes enhanced security features. In response to a request from a user, a teller check image is generated using a check template and that includes check information in fields of the check template. The teller check also includes a unique visible pattern or unique visible picture in the image that serves as an enhanced security feature. A captured image or received image of the teller check is able to be verified based on data extracted from the unique visible pattern or unique visible picture.

Abstract: Smart gear, which includes electrically conductive fabric, is provided to monitor health parameters, conduct transactions, and generate power. For example, the smart gear may include one or more sensors coupled to the fabric and/or electrically conductive fibers woven into the fabric. The one or more sensors may monitor health parameters such as breathing patterns, sleep patterns, heart murmurs, and so forth. The smart gear may include one or more wireless regions that power electronic devices based on inductive charging.

Abstract: Methods and systems described herein are directed to evaluating an individual's ecological impact to promote reductions for that impact. An ecological impact evaluator can automatically retrieve behavioral data of a user that the evaluator can assess for guiding the user with respect to one or more of product purchasing, water usage, recycling and composting procedures, and driving practices. In view of data collected, the evaluator can provide various recognitions for behavior that incentivize the user toward maintaining or improving current behaviors.

Abstract: A system and method for performing machine learning in a mobile computing device which is configured to be coupled with a cloud computing system is disclosed. The method may include activating, on the mobile computing device, a machine learning application, which accesses a local machine learning system including a local machine learning model, periodically updating the local machine learning system based upon updates for the local machine learning system received from a global machine learning system hosted by the cloud computing system, performing machine learning based on received training data, and periodically transmitting changes to the local machine learning system from the mobile computing device to the global machine learning system hosted by the cloud computing system.

Abstract: A safety deposit box facility that allows for 24/7 remote access by authorized personnel to their safety deposit boxes via an on-site robot. The facility includes safety deposit boxes that can be opened along one side by the robot, and on the other side by the authorized person. In some embodiments, the authorized person may request a remote access event in which they can view the contents of their safety deposit box via a camera associated with the robot, and/or request a secure delivery of one or more items disposed in their safety deposit box.

Abstract: A computer system and method having a user interface including a touch-sensitive display screen. The system and method enables entry of a password which includes displaying a first array of a plurality of images on the touch-sensitive display prompting a user to select with a finger one of the plurality of images displayed. Subsequently at least another array of a plurality of images successive to the first array is displayed on the touch sensitive display prompting a user to select with a finger one of the plurality of images displayed in the another array of images. A computer processor then determines if a user selected a predetermined image from the first array of the plurality of images and a predetermined image from each at least another array of plurality of images displayed after the first array. If determined, the user is permitted access to an application executable on the computer system.

Abstract: A system may include a first vehicle and a first control system that may control one or more vehicle operations of the first vehicle. The first control system may perform operations including receiving a first vehicle operation signal from a second control system associated with a second vehicle, sending one or more requests to one or more computing systems for a confirmation of the first vehicle operation signal, and determining a set of vehicle instructions to control the one or more vehicle operations based on whether one or more responses from the one or more computing systems provide the confirmation. The first control system may then control the first vehicle based on the set of vehicle instructions.

Abstract: The present disclosure relates to detecting the use of fake voice command to activate microphones of smart devices. In one embodiment, sound characteristics associated with an audio signal from a microphone of smart device may be compared with other microphones of the smart device in order to detect fake voice commands. In another embodiment, sound characteristics associated with the audio signal from the microphone may be compared with a threshold range of stored sound characteristics in order to detect fake voice commands. In some embodiments, a controller may triangulate a position associated with a source of a sound in order to detect a fake voice command. In a further embodiment, a controller may verify that a user or associated electronic device are near a smart device to authorize a voice command.

Abstract: A system and method for using gaming as a tool to determine optimum routes through areas that have experienced a disaster are provided. Due to the disaster, an area may have some destroyed roads and other infrastructure. In one embodiment, input data collected from people on the ground, drones, and other aerial vehicles can be used to generate a virtual simulation of the area. The virtual area may include renderings of destroyed roads and infrastructure. Once created, the virtual area may be uploaded to an online gaming environment where users can participate in online games that permit them to find the best routes through the virtual area between starting and destination points. Once optimal routes are found, the system may send these routes to navigation systems used by rescuers and other people in the area to help them navigate the disaster area.

Abstract: Systems and methods for fault tolerant digital payments are disclosed. The system comprises a computing system that can automatically determine if payments should be processed using a digital payment processing system, or using a blockchain. When the system detects that communication with the digital payment processing system has been interrupted, the system may facilitate transactions through the blockchain. In particular, the system may receive private keys from mobile devices of customers, which can be used to make payments on the blockchain.

Abstract: The present disclosure is directed to systems and processes for adaptively optimizing query execution. To improve efficiency and reduce costs of executing queries, machine learning may be utilized to automatically choose or adjust the processing blocks used to execute or process queries based on the nature of the cloud processing outputs and client preferences, among other considerations. A machine learning engine may group queries together into topics based on characteristics of the queries. The machine learning engine may receive execution statistics for the topics and, using the execution statistics, identify which topics enhance or optimize query execution. The machine learning engine may determine which topics may be executed by which processing blocks, at which times of day, and on which platforms, among other considerations in order to enhance or optimize cost or processing efficiency according to client-specified parameters.

Abstract: A method and system of providing customer-specific information and guidance to support agents during calls with customers. The information includes a summary of in-app activity by the customer prior to the call between the customer and a support agent. The method includes receiving and storing the in-app activity in an activity record associated with the customer's account. The activity record is retrieved in response to a communication session being initiated or occurring between the customer and support agent and presented to the support agent in order to facilitate the conversation and expedite the resolution process.