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: Disclosed is a method of establishing and managing an inter-institution transaction system. The method comprises creating plurality of threads to constitute parallel fund transfer system, wherein each thread is a replica of single fund transfer system, and wherein plurality of threads are to be operated in parallel; assigning plurality of institutions to plurality of threads, wherein a given institution is assigned as a sending institution and/or a receiving institution of a financial transaction to at least one thread, and a given thread has at least one sending institution-receiving institution pair assigned thereto; assigning financial transactions to plurality of threads based at least on sending institution-receiving institution pairs assigned to plurality of threads; and processing financial transactions at their corresponding threads.

Abstract: Methods and systems for facilitating a two-way user accounts onboarding and linking across multiple service providers are presented. Upon logging in a user device to a first user account of a first service provider, a first user interface associated with the first service provider is provided on the user device. Through the first user interface, a user of the user device may request creating a second user account with a second service provider. The first service provider may then exchange access tokens and refresh tokens with the second service provider to link the first user account with the second user account. After the linking, the first service provider may configure the first user interface to provide the user access to the second user account without requiring the user to log in to the second user account.

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: Disclosed are methods and systems for validating instruments. For instance, a request to generate a cashier's check may be received from an application executing on a drawer's computing device. Request information may be stored within a data storage entry for the check, and a code to be printed on the check may be generated to include at least a portion of the request information. Instructions may be provided to a printing system to cause the check to be printed with the code, where the printed check is invalid. Instructions may be provided to the application and/or printing system to cause display of a prompt to facilitate a reading of the code to validate the check. When information within the code is read by and received from the printing system, the check may be validated, which triggers a transfer of funds from the drawer's account. Otherwise, the check may remain invalid.

Abstract: Techniques are described for determining account features based on a risk assessment. A first set of account features may be determined, including security feature(s) such as mode(s) for authenticating and/or verifying the identity of a user associated with account(s). Based on the first set of features, a risk metric may be determined for the account(s). The risk metric may indicate a risk that fraud may be committed against the account or using the account. Based on the determined risk metric, a second set of account features may be determined for the account(s). The first and second sets of account feature(s) may be applied to the account(s). Disabling a particular feature may cause a reevaluation of the risk metric and a redetermination of the feature sets to be applied to the account(s).

Abstract: There are provided systems and methods for a machine learning model and narrative generator for prohibited transaction detection and compliance. A service provider server, such as an electronic transaction processor, may generate a machine learning model using a supervised training technique, which may detect transactions that may be money laundering. The model may be iteratively trained by detecting flagged transactions and outputting those transactions to an agent for identification of false positives, which may be used to retrain the model. When outputting the flagged transactions, a narrative may be generated using an explainer graph and a machine learning prediction explainer that identifies the features of the transaction data that caused the transactions to be flagged. Further, once the model is trained additional transactions may be processed to determine whether the features of those transactions indicate prohibited behavior.

Abstract: A payment network is operated with a blockchain-based ledger. In one embodiment, a request is prepared to complete a transaction from an account associated with a payer digital wallet for entry on a blockchain. The request includes an amount and a payee identifier associated with a payee digital wallet. The request is sent to the blockchain and approved. A balance of the payer digital wallet and a balance of the payee digital wallet are adjusted in response to approval of the request to complete the transaction by at least writing a portion of data associated with the transaction comprising the amount and the payee identifier to the blockchain. A reputation ledger associated with the account is recorded in the blockchain in association with the transaction. The reputation ledger includes at least one reputation score associated with at least one previous transaction conducted by the account.

Abstract: Through the described contactless payment technology with payment card network to open banking network conversion, the open banking network can be leveraged to bring acceptance for digitized bank account based contactless payments. A converter can receive, via a payment card API channel, a contactless payment message, which includes at least a consumer account number, an acquirer account number, and transaction information for a transaction. The converter can convert the contactless payment message to an open banking message by obtaining a unique B2B trust token corresponding to the consumer account number and creating the open banking message for an open banking API channel using at least the B2B trust token, the acquirer account number, and the transaction information. The converter can communicate, via the open banking API channel, the open banking message to a consumer bank system.

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: An apparatus and method for generating a pecuniary program, the apparatus including a computing device, configured to receive a user input relating to a user; receive pecuniary data relating to a user; identify a plurality of trends in the pecuniary data; generate a first training data set including: at least a priority scoring criteria; and a plurality of a plurality of identified trends in pecuniary data relating to the user; classify at least an element of the user input to a priority score using a first machine learning model, wherein classifying the at an element of the user input includes training a first machine machine-learning model, as a function of the first training data set, and generate a pecuniary program for the user as a function of the priority score.

Abstract: Systems and methods relating to improving the experience of gig-economy workers are disclosed, with particular reference to gig-economy work involving vehicle use. Such systems and methods include automatically monitoring and evaluating activities such as driving during performance of gigs, as well as providing recommendations based thereupon. Gig-related activities may be automatically detected based upon data provided by a mobile device associated with the gig-economy worker, thereby automatically generating a record of such activities for the worker. Telematics data indicative of movement of a vehicle may be collected and analyzed to detect gig-economy activities. During or after performance of gig-economy work, data automatically collected may be used to generate and present recommendations or education points to the gig-economy worker.

Abstract: The present disclosure relates to systems and methods for generating real-time quotes using machine learning algorithms. The system may include a processor in communication with a client device, and a storage medium storing instructions that, when executed, cause the processor to perform operations including: receiving an image of a vehicle from the client device, extracting one or more features from the image, based on the extracted features and using a machine learning algorithm, identifying one or more attributes of the vehicle, based on the identified attributes of the vehicle, determining a make and a model of the vehicle, obtaining comparison information based at least in part on the determined make and model, estimating a quote for the vehicle based on the comparison information; and transmitting the estimated quote for display on the client device.

Abstract: Aspects described herein may allow for detecting fraudulent transaction requests using light sensors on smartcards. A computing device may receive, from a terminal, a request to approve a transaction. The request may comprise a timestamp corresponding to a time when the transaction was requested, a transaction type associated with the transaction, and an identification of a card associated with the transaction. The computing device may also receive, from one or more light sensors associated with the card, light data. The light data indicates an amount of ambient light during a time period when the transaction was requested. The computing device may determine, based on the transaction type, a model of expected ambient light. If the light data corresponds to the model, the computing device may send, based on the determination that the light data corresponds to the model, an approval of the request.

Abstract: In various embodiments, the mobile device includes an NFC tag for near-field communication. The NFC tag includes a first storage device and a second storage device. Personal data is stored on the first storage device, which is not accessible by an NFC reader device. Prior to a transaction, the personal data is transmitted from the first storage device to the second storage device, where the data is readable or accessible by the reader device. In one embodiment, the data remains stored in the first storage device and available for future use. In another embodiment, the data comprises financial product data.

Abstract: A data transaction processing system receives electronic data transaction request messages from client computers over a data communication network and augments each message with hardware level data, and generates a monotonically increasing identification number for each electronic data transaction request message based on the hardware level data. The data transaction processing system transmits the identification number to the client computer utilizing transport layer protocols.

Abstract: Disclosed is an example approach in which network and non-network features are used to train a predictive machine learning model that is implemented to predict financial crime and fraud. Graphical network features may be generated by applying financial entity risk vectors to a network model with representations of various types of networks. The network model may comprise transactional, non-social, and/or social networks, with edges corresponding to linkages that may be weighted according to various characteristics (such as frequency, amount, type, recency, etc.). The graphical network features may be fed to the predictive model to generate a likelihood and/or prediction with respect to a financial crime. A perceptible alert is generated on one or more computing devices if a financial crime is predicted or deemed sufficiently likely. The alert may identify a subset of the set of financial entities involved in the financial crime and present graphical representations of networks and linkages.

Abstract: There are provided systems and methods for a machine learning-based network trained to perform risk assessment through device data. A service provider server receives device data of a user device associated with a merchant account registered with a merchant server for a merchant service, and receives, from the merchant server, a request containing a unique token identifier for initiating a risk assessment operation to generate a device assessment score for the user device, where the device assessment score indicates a level of risk between the user device and the merchant account. The service provider server selects a risk assessment engine to perform the risk assessment operation and generates the device assessment score and a narrative for the device assessment score, and sends, to the merchant server through an application programming interface, a message containing a unique device identifier for the user device, the device assessment score and the narrative.

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 regularization-based (RB) hedging tool identifies a recommended hedging portfolio that track a target asset and provides one or metrics indicating a predicted performance of the hedging portfolio relative to the target asset. The RB hedging tool uses a RB hedging model that is trained on price data from an observation period. Initial hyperparameters for the model are selected using asset price data from a validation period and the performance of the model is evaluated by applying it to asset price data from a backtest period. The end-user is presented with one or more metrics indicating the performance of the model and may modify one or more settings (e.g., hyperparameters) of the model. The model is retrained and reapplied to the backtest period, and the metrics are updated. Thus, end-users may tailor the model to their own particular needs and preferences.