Abstract: Techniques for training a classification model to improve the classification of open banking transactions are presented. The techniques include receiving raw training data from a data source. The raw training data includes historical transaction data made up of a plurality of individual transactions. The raw training data is input into the classification model. The raw training data is processed by performing a data preparation operation on the raw training data. The data preparation operation includes removing numerical characters, repeating special characters, and accent words from the textual data of each transaction. Vocabulary training is then performed on the processed training data, including tokenizing the text of each transaction and converting the tokenized text into a transformer model specific format. The classification model is then trained using a transformer model, which uses the tokenized text. The trained classification model is then stored in a database.

Abstract: Techniques for training an entity resolution model are presented. The techniques include receiving a minimum viable data product (MVDP) scope, a product scope, and a data mining scope from a user. A data mining goal is determined based on the MVDP scope, product scope, and data mining scope. One or more proof of concept (PoC) models are defined based on the data mining goal, and one of the PoC models is selected for training. A trained deep learning model is generated by iteratively training the selected PoC model. The trained deep learning model is then tested and validated against a predefined achievable loss metric using a sample labelled dataset for testing.

Abstract: Systems and methods for reducing false positives for financial transaction fraud monitoring using machine learning techniques. Using an original model for separating transactions into high risk and low risk categories for fraud, transactions falling into the high-risk category may be labeled as a false positive or a true positive. The labels and data associated with the transactions may be used to train two or more false positive reduction models (FPRMs) using iterative machine learning techniques. Once training is complete, a future transaction may be processed using the original model, and, if the original model indicates that the future transaction is high risk, data associated with the future transaction may be processed by the trained FPRM(s), which may determine whether the future transaction is at a high risk or a low risk of being fraudulent.

Abstract: A system and computer-implemented method for identifying fraudulent healthcare providers receives raw claims data from one or more data sources. The raw claims data includes claims associated with a selected healthcare provider. Each of the claims includes one or more claim lines. A first model is executed on the raw claims data. The first model determines a first score for the healthcare provider. A second model is executed on the raw claims data. The second model determines a second score for the healthcare provider. In addition, a third model is executed on the raw claims data. The third model determines a third score for the healthcare provider. A final provider-level risk score is determined for the healthcare provider based on the first, second, and third scores.

Abstract: Techniques for training an entity resolution model are presented. The techniques include inputting raw training data into the entity resolution model. The training data includes historical transaction data including a plurality of transactions. A label dictionary is generated by performing natural language processing (NLP) on the training data. The NLP includes scanning text of each transaction, extracting one or more entities from the text, and storing the label dictionary in a database. The label dictionary includes the extracted entities. Tagged data is generated from the training data using the label dictionary. Vocabulary training is performed on the training data, including tokenizing the text of each transaction and converting the tokenized text into a transformer model specific format. The entity resolution model is then trained using a transformer model, which uses the tokenized text and the tagged data. The trained entity resolution model is then stored in a database.

Abstract: A swap check oracle receives a transfer request from a user or smart contract on a first blockchain indicating a first digital asset to be transferred. The swap check oracle verifies the authenticity of the user and/or digital asset and instructs the smart contract to transfer the first digital asset to a custodial blockchain address on the first blockchain. Another swap check oracle performs the same process for a second digital asset from a second user on a second blockchain. A central processing server is notified of the successful transfer of the digital assets to the custodial addresses on both blockchains, verifies the holding of the digital assets by the custodial addresses, and then initiates a release of the digital assets to the new parties on both of the blockchains.

Abstract: A method for optimized validation of a blockchain transaction through the use of a parallel database and date management includes: receiving a new blockchain transaction that includes unspent transaction outputs; identifying, in a parallel database, a start date for a blockchain wallet involved in the new transaction; identifying a subset of blocks in the blockchain that have been added to the blockchain since the start date; validating the new blockchain transaction by validating the unspent transaction outputs using the subset of blocks; and, if the validation is successful, adding the new transaction to the blockchain, or, if the validation is unsuccessful, notifying the submitter of the failed validation.

Abstract: An automatic billing updater (ABU) manager computing device for managing an account information data source and tracking access to the account information data source is provided. The ABU manager computing device receives account data from an issuing party, stores the account information in an account information data source, and provides the account data to requesting parties. The ABU manager computing device tracks access to the stored account data via an ABU traffic data source. ABU manager computing device may also track transaction-related messages corresponding to the stored account data in a transaction traffic data source. In certain embodiments, ABU manager computing device may generate and transmit notifications based on data stored in one or more of the stored account data, the ABU traffic data source, and the transaction traffic data source.

Abstract: A computer system includes a processor and a memory device. The computer system is configured to execute a function that builds a software instance definition object. The software instance definition object includes one or more configuration items, attributes, permissions, and linkages to other objects, which define a software instance. The computer system temporarily stores the software instance definition object in the memory device. The computer system updates at least one of the configuration items, attributes, permissions, and linkages to other objects of the stored software instance definition object, thereby creating an updated software instance definition object. A fix command is then executed to update the software instance based on the updated software instance definition object, thereby creating an updated software instance. After creating the updated software instance, the software instance definition object is deleted from the memory device.

Abstract: A system and method for transmitting real-time messages within a supply chain financing (SCF) network using a using a customized and improved messaging protocol is provided. A computing device receives a SCF request message having an SCF messaging protocol including a plurality of data fields including SCF request data related to a request for supply chain financing including an early payment identifier, parses the SCF request message based upon the SCF messaging protocol, selects a financing entity for receiving the request, stores within the memory device the SCF request data, transmits an early payment message having the same messaging protocol to the selected financing entity, receives a confirmation message having the same SCF messaging protocol from the financing entity including a plurality of data fields including SCF confirmation data, and stores within the memory device the SCF confirmation data with the SCF request data.

Abstract: A system and method for generating and provisioning payment credentials to a mobile device lacking a secure element includes receiving and storing by the mobile device a card profile from a remote system. The card profile may include payment credentials corresponding to a payment account and a profile identifier. The mobile device may receive a mobile personal identification number (PIN) input by a user of the mobile device and transmit a key request to the remote system. The mobile device may receive a single use key which may include an application transaction counter and a generating key from the remote system. The mobile device may generate a payment cryptogram valid for a single financial transaction based on the received single use key and the mobile PIN and transmit the payment credentials and the generated payment cryptogram to a point-of-sale terminal for use in a financial transaction.

Abstract: The disclosure herein relates to methods and systems of that facilitate instant payment of invoices based on automatic, intelligent, predictions that a buyer will approve an invoice without express input by the buyer. For example, a system may determine a probability that an invoice will be approved for payment by a buyer with express approval from the buyer. Once an invoice is automatically approved for payment, the system may employ various automated systems to facilitate payment to suppliers without the need for suppliers to extract virtual card data or other payment information from email payment requests or other invoice payment requirements to receive payment. For example, the system may generate calls to automatically generate virtual card numbers on behalf of buyers and submit payment messages for payments to suppliers using the virtual card numbers.

Abstract: A system, method, and computer-readable storage medium configured to facilitate cash transactions at an Automated Teller Machine when an ATM card is not present. The system allows a customer of a first bank to use an ATM of a different bank and still facilitate the ATM transaction without an ATM card present.

Abstract: A method for verification of a data value via a Merkle root includes: storing, in a memory of a processing server, a Merkle root; receiving at least a data value, a nonce, and a plurality of hash path values; generating a combined value by combining the data value and the nonce; generating a first hash value via application of a hashing algorithm to the combined value; generating a subsequent hash value via application of the hashing algorithm to a combination of the first hash value and a first of the plurality of hash path values; repeating generation of the subsequent hash value using a combination of the next hash path value of the plurality of hash path values and the most recent subsequent hash value; and verifying the data value based on a comparison of the Merkle root and the last generated subsequent hash value.

Abstract: A system and method for authenticating a location of a payment acceptance device is provided. The method comprises receiving a transaction request comprising merchant data, the merchant data identifying the payment acceptance device and location information identifying a location at which the transaction request is initiated; retrieving, from a database, a payment acceptance device identifier based on the merchant data, the payment acceptance device identifier identifying a location at which the payment acceptance device is registered; authenticating the location of the payment acceptance device by comparing the location information and the location identified by the payment acceptance device identifier.

Abstract: A method and system for detecting fraudulent network events in a payment card network by incorporating breach velocities into fraud scoring models are provided. A potential compromise event is detected, and payment cards that transacted at a compromised entity associated with the potential compromise event are identified. Subsequent transaction activity for the payment cards is reviewed, and a data structure for the payment cards are generated. The data structure sorts subsequent transaction activity into fraud score range stripes. The data structure is parsed over a plurality of time periods, and at least one cumulative metric is calculated for each of the time periods in each fraud score range stripe. A plurality of ratio striping values are determined, and a set of feature inputs is generated using the ratio striping values. The feature inputs are applied to a scoring model used to score future real-time transactions initiated using the payment cards.

Abstract: A growth predictor includes a monitor, a prediction engine, and a prioritization engine. The monitor receives or generates first information of a network already identified as a candidate money laundering (ML) network by an anti-money-laundering system. The prediction engine predicts second information indicative of a growth size of the ML network at a future time based on the first information. The prediction engine executes one or more predictive models to generate the second information indicative of growth size based on the first information, which indicates one or more changes that have occurred in the candidate ML network over a past period of time. The prioritization engine determines a priority of the candidate ML network based on the second information.

Abstract: A method for enabling real-time authorization of a transaction with pooled equity includes: receiving a transaction request message, the transaction request message including at least a public key of a cryptographic key pair and a transacting currency amount; verifying a balance of a cryptographic currency for a first blockchain wallet associated with the cryptographic key pair on a blockchain using at least the public key, where at least a portion of the balance is locked in a currency pool; generating a smart contract, the smart contract configured to initiate a first blockchain transaction on the blockchain for transfer of the transacting currency amount from a second blockchain wallet to a destination address, where the second blockchain wallet is different from the first blockchain wallet; and transmitting the generated smart contract to a node in a blockchain network associated with the blockchain for publishing to the blockchain.

Abstract: A cardholder authentication method includes receiving, at an authentication network, an authentication request involving an account. The method further includes determining, based at least in part on a portion of an account identifier associated with said account, an authentication service. In addition, the method includes determining, based at least on said authentication service and a portion of said account identifier, an authentication response. The method also includes transmitting, to a merchant associated with a transaction involving said account, said authentication response.

Abstract: Embodiments provide methods and systems for verifying plurality of user attributes corresponding to a purchase of a user, including restricted products. Methods include receiving a payment request for purchasing a plurality of products, determining plurality of attributes to be verified for the payment request by applying regulatory rules on a products related information associated with the plurality of products, transmitting a payment link, to the merchant terminal including at least one user-specific attribute to be verified from the user.