Abstract: Apparatus, systems, and methods are disclosed that operate to receiving an authentication request at a server associated with an authenticating entity from a requesting party responsive to a request being provided to the requesting party by a client terminal associated with an unauthenticated individual purporting to be an individual account owner previously authenticated with the authenticating entity. A token from the client terminal associated with the unauthenticated individual is received, and the token includes information associated with the unauthenticated individual and a user permission authorizing the authenticating entity to share a selected portion of the information with a plurality of selected requesting parties. The server associated with the authenticating entity authenticates the unauthenticated individual as the individual account owner based on, inter alia, matching the token to a pre-registered identity uniquely associated with the individual account owner.

Abstract: Various embodiments herein each include at least one of systems, devices, methods, and software for automated session closing upon location-sensed departure. Some such embodiments may be implemented at retail outlets (e.g., restaurants, salons, stores, etc.) to automatically charge a customer account for ordered items upon an automated detection of a customer retail outlet departure. One embodiment, in the form of a method performed on a server-based system, includes receiving, via a network from a mobile device associated with a customer account, first data indicating the mobile device has departed a retail outlet. The method then generates, from stored retail session data, a final bill including an amount payable and then charges an amount payable from the final bill to a payment account associated with the customer account.

Abstract: Methods, systems and computer-readable storage media are provided for controlling operation of a trading algorithm based on operating condition rules. Certain embodiments provide a method including determining, using a computing device, an approval of use of a trading algorithm by monitoring for an occurrence of an operating condition defined in an operating condition rule. The example method includes determining if the trading algorithm complies with the operating condition rule during the occurrence of the operating condition. The example method includes sending, using the computing device, a notification to a trading instrument to approve or not approve the use of the trading algorithm. The trading algorithm is to be used to implement a trading strategy.

Abstract: A system configured to: (a) retrieve structured and unstructured data from one or more external data sources, the structured data including time-series data on a financial instrument and the unstructured data including words; (b) analyze the unstructured data to determine a sentiment measure for the financial instrument; (c) analyze the structured data to obtain a training dataset; (d) train a neural network model with the training dataset such that the neural network can provide a predicted price of the financial instrument for a future timestamp; and (e) provide a decision for managing the financial instrument based at least in part on the sentiment measure for the financial instrument, the predicted price of the financial instrument, and a current holding of the financial instrument.

Abstract: There are provided systems and methods for measuring tap pressure on mobile devices to automate actions. A computing device, such as a mobile smart phone, may include a touch screen interface that may accept touch inputs. A touch input may include a presence and location of the touch input on the touch screen interface. Additionally, the computing device may detect a pressure or force applied to the touch screen interface with the touch input. The computing device may include a payment application, where the touch input may be received during use of the payment application. The pressure of the touch input may further define the input for the payment application. For example, a touch input with a first pressure may cause a first process to be initiated, while a second pressure may cause a second process to be initiated.

Abstract: A credit card recommendation system for recommending credit cards to a user can be based on the consumer's estimated monthly spend, estimated spend across a plurality of categories, and user credit data. The credit card recommendation system can filter credit cards based on a likelihood of approval for the user. The credit card recommendation can determine a reward valuation and an adjustment valuation by assessing user spend and the characteristics corresponding to the credit card. The credit card recommendation can train a model to score credit cards for users, and apply specific user data to the model to determine a credit card score particular to the user.

Abstract: A computer-implemented method includes receiving, by a mobile wallet computing system, information relating to a usable object from a party and provisioning, by the mobile wallet computing system, the usable object to a mobile wallet associated with a user. The computer-implemented method further includes receiving, by the mobile wallet computing system, one or more rules relating to the usable object from the party, wherein the one or more rules define an activation condition for the usable object such that when the activation condition is fulfilled the usable object is active in the mobile wallet of the user; determining, by the mobile wallet computing system, whether the activation condition is fulfilled; and in response to the activation condition for the usable object being fulfilled, activating, by the mobile wallet computing system, the usable object in the mobile wallet of the user.

Abstract: Techniques are disclosed relating to determining a risk score for domains associated with a transaction. In some embodiments, a transaction computer system receives transaction details for a transaction between a consumer and a merchant, where the transaction details are received from the merchant real-time with the transaction and include a set of transaction URLs for subsequent use in the transaction. The computer system may receive, from a browser of the consumer that is used to initiate the transaction, URL referrer information real-time with the transaction, where the URL referrer information indicates a referring web page to the transaction computer system. The computer system may determine, using the set of transaction URLs and the URL referrer information, a set of domains for the transaction and then determine a risk score for the set of domains. The computer system may determine, based on the risk score, whether to allow the transaction.

Abstract: A method is disclosed. The method includes obtaining an access request associated with a user for a software application; obtaining a plurality of verification attributes associated with the user; generating a fraud score for the access request by feeding a supervised machine learning (ML) classifier with a feature vector for the user that is based on the plurality of verification attributes; selecting a first unsupervised ML anomaly detector of a plurality of unsupervised ML anomaly detectors based on the fraud score; generating an anomaly score for the access request by feeding the first unsupervised ML anomaly detector with an augmented feature vector for the user that is based on the plurality of verification attributes and the fraud score; and processing the access request based on the anomaly score.

Abstract: System and methods for pre-marshalling messages in an electronic trading environment are provided. Specifically, pre-marshalling messages allows for client and network devices to create messages in response to a defined first condition. Pre-marshalling messages allows a trading system to do more processing up front before the message is needed and when a second condition is satisfied. Thus, when the message is needed no further or very little processing must be performed to generate and send the message. The client and/or network device may allocate a separate, unused, or under-utilized processing thread to the task of creating pre-marshalled messages. Pre-marshalled messages may be stored in memory at the network device until it is determined that the second condition has been satisfied. Once a pre-marshalled message is sent, the unused pre-marshalled messages that were stored in memory may be deleted or may be overwritten with newly pre-marshalled messages.

Abstract: Systems, methods, and computer program products to provide credit decisioning based on graph neural networks. A lending network graph of a plurality of loans may be received, each loan associated with a creditor and one account. A first node of the graph may be associated with a first creditor and the second node may be associated with a first account. A graph neural network may receive a respective message from each node connected to the first node, each message comprising an embedding vector reflecting a current state of the node. The graph neural network may update weights for the first node in a forward pass. The graph neural network may receive a respective message from each node connected to the second node, each message comprising the embedding vector reflecting the current state of the node. The graph neural network may update weights for the second node in a backward pass.

Abstract: This application describes apparatus and methods for distributing computing resources using edge-computing. Apparatus may include an edge node that is positioned in a target geographic region. The edge-node may include information needed to process transactions that occur locally in the target geographic region. The edge-node may process the transaction without communicating with a central server. The edge-node may support processing transactions on behalf of a variety of financial institutions and merchant processing systems. Processing the transaction locally may reduce computational resources typically required by region server to process high volumes of transactions. The edge-nodes would process transactions locally, close to a location of at least one party to the transaction.

Abstract: Systems and methods for facilitating at least a portion of a secure electronic financial transaction are provided herein. Methods may include generating a unique payee identifier that represents a unique payment relationship between a payee and one or more payors, receiving a transaction request, the transaction request comprising the unique payee identifier and a payment, and authorizing a payment to the payee that corresponds to the payment.

Abstract: The disclosed embodiments relate to ensuring that a selected value, selected, for example, via interaction with a graphic user interface, of a dynamically changing parameter, such as a price, is used when generating an electronic data transaction request message in a data transaction processing system, such as an electronic trading system. The data transaction processing system being a system in which data items, such as financial contracts, e.g. futures contracts, are transacted or otherwise traded by a hardware matching processor that attempts to match electronic data transaction request messages with electronic data transaction request messages counter thereto for the same one of the data items based on multiple transaction parameters. A selected value is temporarily buffered allowing the represented value to update wherein a subsequently generated transaction may be based on the buffered value rather than an updated value.

Abstract: Aspects of the disclosure relate to using machine learning techniques for generating automated suspicious activity reports (SAR). A computing platform may generate a labelled transaction history dataset by combining historical transaction data with historical report information. The computing platform may train a convolutional neural network using the labelled transaction history dataset. The computing platform may receive new transaction data and compress the new transaction data using lossy compression. The computing platform may input the compressed transaction data into the convolutional neural network, which may cause the convolutional neural network to output a suspicious event probability score based on the compressed transaction data.

Abstract: A computer-implemented system and method receive information regarding a funds transfer from a payor to a payee. The information is received by a depository computer system. The depository computer system forwards the information regarding the funds transfer to an account verification service computer system. The depository receives an indication of a real time account status of an account of the payor. The account status information is received from a paying bank computer system that maintains the account of the payor via the account verification service computer system. Release of the funds to the payee is authorized based on the account status information received via the account verification service computer system. The authorization occurs in real time relative to when the funds transfer information is initially received.

Abstract: Methods, apparatus, and processor-readable storage media for automatic processing of claims using artificial intelligence are provided herein. An example computer-implemented method includes obtaining at least one image related to at least one device damage claim, automatically classifying the at least one device damage claim by applying one or more artificial intelligence techniques to the at least one obtained image, and performing one or more automated actions based at least in part on the classification of the at least one device damage claim.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for evaluating risk are provided. One of the methods includes: obtaining a plurality of transactions each comprising a plurality of data dimensions, wherein some of the plurality of transactions are labeled as risky transactions and some of the plurality of transactions are labeled as safe transactions; obtaining at least one of the plurality of data dimensions as an output space and the plurality of data dimensions other than the at least one data dimension as an input space; initializing a first mapping from the input space to a latent space and a second mapping from the latent space to the output space, wherein the first mapping comprises mapping the input space to the latent space according to an inverse of a generalized linear model; and optimizing the first mapping and the second mapping to generate a Bezier surface.

Abstract: A device and method for assisting a user in photographing a vehicle for documenting a condition of the vehicle. The method includes receiving information regarding the vehicle and displaying a graphic overlay of the vehicle. A video image of the vehicle is captured with a camera and is displayed with the graphical overlay. The video image of the vehicle is aligned with the graphical overlay of the vehicle and the image of the vehicle is stored. The graphical overlay can be oriented on the screen in a manner so that the stored image of the vehicle is captured at a predetermined horizontal angle, vertical height and/or vehicle size. The photograph of the vehicle can be attached to an insurance estimate. Standardizing the horizontal angle, vertical and/or vehicle size of the photographed vehicle promotes consistency and uniformity in the estimating process.

Abstract: Logic may detect fraudulent transactions. Logic may determine, by a neural network based on the data about a transaction, a deviation of the transaction from a range of purchases predicted for the customer, wherein the neural network is pretrained to predict purchases by the customer based on a purchase history of the customer. Logic may compare the deviation of the transaction from purchases predicted by the customer against a deviation threshold to determine whether the transaction is within the range of purchases predicted by the neural network. Logic may generate a notification in response to a determination that the deviation of the transaction from the range of purchases predicted exceeds a deviation threshold, the notification to identify the transaction as a potentially fraudulent transaction. Logic may train the neural network based on the transaction in response to a determination that transaction is not a fraudulent transaction.