Mastercard International Incorporated patent applications published on November 30th, 2023

Patent applications for Mastercard International Incorporated on November 30th, 2023

METHOD AND SYSTEM FOR MAINTAINING PRIVACY AND COMPLIANCE IN THE USE OF ACCOUNT REISSUANCE DATA (18232408)

Main Inventor

Andrew REISKIND

Brief explanation

The patent application describes a method for linking payment accounts.

• The method involves storing multiple account profiles, each containing data related to a payment account.
• The account number in each profile is encrypted using a specific encryption method.
• Account linkage data is received, which includes encrypted account identifiers that are indicated as being linked to each other.
• The encrypted account identifiers are matched to the encrypted account numbers.
• One or more account profiles are updated to indicate a link to another profile if the encrypted account number in the profile matches an encrypted account identifier that is linked to another encrypted account identifier that matches the encrypted account number in the other profile.

Abstract

A method for linking payment accounts includes: storing a plurality of account profiles, each profile including data related to a payment account including an account number and account data; encrypting the account number included in each account profile using a method of encryption to obtain an encrypted account number; receiving account linkage data, the data including a plurality of encrypted account identifiers, each identifier being indicated as being linked to another identifier, and each identifier being encrypted using the method of encryption; matching each of the encrypted account identifiers to an encrypted account number; and updating one or more account profiles to indicate a link to another account profile where the encrypted account number included in the profile being updated matches an encrypted account identifier that is indicated as being linked to an encrypted account identifier that matches the encrypted account number included in the other profile.

ARTIFICIAL INTELLIGENCE ENGINE FOR TRANSACTION CATEGORIZATION AND CLASSIFICATION (18324307)

Main Inventor

Yogesh Sakpal

Brief explanation

- The patent application describes techniques for improving the classification of open banking transactions using a classification model.

- The techniques involve receiving raw training data, which includes historical transaction data. - The raw training data is processed by performing a data preparation operation, which involves 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 it into a specific format for a transformer model. - The classification model is trained using a transformer model that utilizes the tokenized text. - The trained classification model is stored in a database.

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.

ARTIFICIAL INTELLIGENCE ENGINE FOR ENTITY RESOLUTION AND STANDARDIZATION (18324315)

Main Inventor

Yogesh Sakpal

Brief explanation

- The patent application describes techniques for training an entity resolution model.

- The entity resolution model is trained using raw training data, which includes historical transaction data. - A label dictionary is generated by performing natural language processing (NLP) on the training data. - The label dictionary includes extracted entities from the text of each transaction. - Tagged data is generated from the training data using the label dictionary. - Vocabulary training is performed on the training data by tokenizing the text of each transaction and converting it into a transformer model specific format. - The entity resolution model is trained using a transformer model, which utilizes the tokenized text and the tagged data. - The trained entity resolution model is stored in a database for future use.

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.

SYSTEM AND METHOD OF TOKENIZING DEPOSIT ACCOUNT NUMBERS FOR USE AT PAYMENT CARD ACCEPTANCE POINT (18448587)

Main Inventor

Sandeep Malhotra

Brief explanation

- The patent application describes a method of routing a payment transaction.

- The method involves receiving a transaction request message that includes a payment token previously issued to an account holder. - The payment token is then translated into a funding account indicator, which is in a format defined for payment card account numbers in a payment account system. - The funding account indicator is further translated into a bank account number that identifies a bank deposit account owned by the account holder. - An Electronic Funds Transfer (EFT) message is then transmitted to initiate an EFT transaction, which will be funded from the bank deposit account owned by the account holder. - The EFT message includes the bank account number.

Abstract

A method of routing a payment transaction includes receiving a transaction request message. The message includes a payment token that was previously issued to an account holder. The payment token is translated into a funding account indicator. The funding account indicator is in a format defined for payment card account numbers in a payment account system. The funding account indicator is translated into a bank account number. The bank account number identifies a bank deposit account owned by the account holder. An EFT message is transmitted to initiate an EFT transaction to be funded from the bank deposit account owned by the account holder. The EFT message includes the bank account number.

SYSTEM AND METHOD OF PAYMENT OF MERCHANTS ON BEHALF OF PAYMENT CARD SYSTEM TRANSACTION ACQUIRERS (18449948)

Main Inventor

Sandeep Malhotra

Brief explanation

The patent application describes methods and systems for handling transactions in a payment card account system. Here is a simplified explanation of the abstract:

• A merchant payment services computer receives and stores the bank account details of a merchant supported by a financial institution.
• The computer also receives card account transaction information for the merchant over a specific period of time.
• Based on the transaction information, the computer calculates the net position for the merchant.
• The computer retrieves the merchant's bank account details from storage.
• An electronic funds transfer is initiated with an originating payment services provider computer to credit the merchant's deposit account with an amount equal to the net position.

Bullet points to explain the patent/innovation:

• The patent application introduces a system for handling payment card transactions for merchants.
• It allows the merchant's bank account details to be securely stored and accessed when needed.
• The system calculates the net position of the merchant based on their card account transaction information.
• An electronic funds transfer is then initiated to credit the merchant's deposit account with the calculated amount.
• This innovation streamlines the process of transferring funds to merchants, ensuring accurate and timely payments.

Abstract

Methods and systems for handling a payment card account system transaction. In an embodiment, a merchant payment services computer receives merchant bank account details for at least one merchant supported by an acquirer financial institution and stores the merchant bank account details in a storage device. The process also includes the merchant payment services computer receiving, from a card payment network computer, card account transaction information for a merchant over a predetermined period of time; calculating, based on the card account transaction information, a net position for the merchant; retrieving merchant bank account details of the merchant from the storage device; and initiating an electronic funds transfer (EFT) with an originating payment services provider (PSP) computer of an electronic funds transfer system to credit the merchant's deposit account in an amount equal to the merchant's net position.

METHOD AND SYSTEM FOR PROCESSING AN ASSET SWAP ACROSS TWO BLOCKCHAINS (17752318)

Main Inventor

Saravana Perumal SHANMUGAM

Brief explanation

- The patent application describes a system that facilitates the transfer of digital assets between different blockchains.

- A swap check oracle is used to verify the authenticity of the user and the digital asset to be transferred. - The first digital asset is transferred to a custodial blockchain address on the first blockchain. - A similar process is followed for a second digital asset from a second user on a second blockchain. - A central processing server is notified once the digital assets are successfully transferred to the custodial addresses on both blockchains. - The central processing server verifies the holding of the digital assets by the custodial addresses. - After verification, the central processing server initiates the release of the digital assets to the new parties on both blockchains.

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.

AGILE ITERATION FOR DATA MINING USING ARTIFICIAL INTELLIGENCE (18324321)

Main Inventor

Siddhesh Dongare

Brief explanation

- The patent application describes techniques for training an entity resolution model.

- The techniques involve receiving input from a user, including a minimum viable data product (MVDP) scope, a product scope, and a data mining scope. - Based on this input, a data mining goal is determined. - One or more proof of concept (PoC) models are defined based on the data mining goal. - One of the PoC models is selected for training. - The selected PoC model is iteratively trained to generate a trained deep learning model. - The trained deep learning model is then tested and validated against a predefined achievable loss metric. - A sample labelled dataset is used for testing the trained model. - The techniques aim to improve the accuracy and effectiveness of entity resolution models.

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.

CONSERVING COMPUTING RESOURCES DURING IDENTITY VALIDATION VIA A LAST USED ACCOUNT (18190086)

Main Inventor

Michael D. McCARTHY

Brief explanation

- The patent application is about improving consumer identity validation (CIV) in payment infrastructure.

- It allows consumers to validate their identity more efficiently by using their last used payment card or account. - This eliminates the need for consumers to go through the entire validation process for each transaction at the same merchant with the same card. - This reduces strain on the payment infrastructure and prevents degradation. - It allows for the allocation of fewer computing resources to CIV, making it easier to deploy and maintain low-power or constrained equipment. - This improves scalability and efficiency of individual CIV performance. - It reduces the overall computing resources and energy required for CIV across the infrastructure. - It enables more CIV transactions to be processed in parallel, reducing processing delays, improving stability, and reducing error rates.

Abstract

Examples provide consumer identity validation (CIV) via last used payment card, improving management of payment infrastructure resources. Examples enable the consumer to more efficiently complete CIV via at least one of a last used account or a last used card. By not requiring consumers to repeatedly go through the entire CIV process at each transaction at the same merchant when using the exact same card they used on a previous visit, excessive and unnecessary strain on the infrastructure is reduced, preventing degradation of the infrastructure. This enables allocating fewer computing resources to CIV, making deployments utilizing low-power or otherwise constrained equipment simpler and easier to maintain. Thus, scaling is enhanced. Individual CIV performance is more efficient, and less computing resources and energy are required across the entire infrastructure. A greater number of CIV transactions are executable in parallel, reducing transaction processing delays, bolstering stability, and reducing error rates.

METHODS AND SYSTEMS FOR REDUCING FALSE POSITIVES FOR FINANCIAL TRANSACTION FRAUD MONITORING USING ARTIFICIAL INTELLIGENCE (17828970)

Main Inventor

Fariborz Nadi

Brief explanation

- The patent application is about reducing false positives for financial transaction fraud monitoring using machine learning techniques.

- It introduces an original model for categorizing transactions into high risk and low risk for fraud. - Transactions labeled as high risk may be either false positives or true positives. - The application proposes training two or more false positive reduction models (FPRMs) using iterative machine learning techniques. - These FPRMs are trained using the labels and data associated with the transactions. - Once training is complete, future transactions can be processed using the original model. - If the original model indicates a future transaction as high risk, the trained FPRMs can process the associated data to determine if it is at high risk or low risk of being fraudulent.

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.

===SYSTEMS AND METHODS FOR DETECTING OUT-OF-PATTERN TRANSACTIONS ([[US Patent Application 18449489. SYSTEMS AND METHODS FOR DETECTING OUT-OF-PATTERN TRANSACTIONS simplified abstract|18449489]])===

Main Inventor

Christopher John Merz

Brief explanation

The patent application describes an inverse recommender system for detecting unusual payment transactions.

• The system includes a memory device and a processor that receives transaction data.
• The transaction data represents past payment transactions between account holders and merchants.
• The processor generates a merchant correspondence matrix, which shows the number of historical payment transactions between pairs of merchants and account holders.
• The merchant correspondence matrix is stored in the memory device, linking the merchant pairs to each account holder.
• When a new payment transaction occurs between an account holder and a merchant, the processor calculates an inverse recommender score based on the account holder's historical payment transaction data.
• The account holder's historical payment transaction data includes information about the merchants they have visited in the past.

Abstract

An inverse recommender system for detecting out-of-pattern payment transactions includes a memory device and a processor programmed to receive transaction data. The transaction data corresponds to historical payment transactions between account holders and merchants. The processor is programmed to generate a merchant correspondence matrix including the merchants and counters indicating the number of historical payment transactions between merchant pairs of the merchants and the account holders. The processor is programmed to store the merchant correspondence matrix in a memory device linking the merchant pairs to each account holder. The processor receives additional transaction data associated with a new payment transaction between an account holder and a merchant, and to generate an inverse recommender score for the new payment transaction based on the account holder's historical payment transaction data. The account holder's historical payment transaction data includes historical payment transaction data associated with the merchants visited by the account holder.

IDENTIFICATION OF FRAUDULENT HEALTHCARE PROVIDERS THROUGH MULTIPRONGED AI MODELING (17828945)

Main Inventor

Athena Stacy-Nieto

Brief explanation

The patent application describes a system and method for identifying fraudulent healthcare providers using raw claims data.

• The system receives raw claims data from various sources, which includes claims associated with a selected healthcare provider.
• The claims data consists of claim lines, which are analyzed using multiple models.
• The first model calculates a score for the healthcare provider based on the raw claims data.
• The second model also calculates a score for the healthcare provider using the same raw claims data.
• The third model determines a third score for the healthcare provider based on the raw claims data.
• Finally, a final provider-level risk score is determined for the healthcare provider by combining the scores from the first, second, and third models.

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.

SYSTEM, METHOD, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIA FOR RECOMMENDING MERCHANTS (18449500)

Main Inventor

Christopher J. Merz

Brief explanation

- The patent application describes a computer system that recommends merchants to cardholders.

- The system receives transaction information from a payment network, including data on purchases made by cardholders at different merchants. - The system also takes into account the preferences of the cardholders for specific merchants. - It determines a rank for each merchant based on the transaction information and cardholder preferences. - Additionally, it determines a neutral rank for each merchant based on the transaction information and the preferences of all cardholders. - The system calculates a score for each merchant by comparing its rank to the neutral rank.

Abstract

A computer system for recommending merchants to a candidate cardholder is provided. The computer system includes a memory device and a processor. The processor receives transaction information for a plurality of cardholders from a payment network. The transaction information includes data relating to purchases made by the cardholders at a plurality of merchants, where the purchases satisfy a first criteria. The processor also receives candidate cardholder preference information for at least one of the merchants input by the candidate cardholder. The processor further determines a merchant rank for each merchant based on the received transaction information and the candidate cardholder preference information, and determines a neutral merchant rank for each merchant based on the received transaction information and neutral cardholder preferences of the plurality of cardholders. The processor also determines a merchant score for each of the plurality of merchants by comparing the merchant rank to the neutral merchant rank.