Mastercard International Incorporated

Po Hu

Jean-Pierre Gerard

Qian Wang

Ramamohan R Sangasani

Henry Weinberger

Shen Xi Meng

Justin X Howe

Tong Zhang

Jean Pierre Gerard

Bruce William Mac Nair

Rohit Chauhan

Serge Bernard

Anant Nambiar

Marc Del Bene

Fuyuan Xiao

Daniel G Salazar

Bruce W Mac Nair

Disclosed herein are systems and methods of individual level learning that include receiving purchase event data from a merchant device that indicates that a purchase event occurred by a user on a user device, and transmitting the purchase event data to an analytics server. The methods may also include processing the purchase event data. The processing may include calculating a time gap for each of two sequential purchase events in a list of purchase events, and calculating an average duration of consecutive events by averaging all of the purchase events in the list of purchase events. The method may determine a purchase hazard probability that a purchase event will occur on a given day, when the average duration of consecutive events is larger than a standard deviation of the event occurring. When the purchase hazard probability is above a threshold, the system may push a message to the user device.

Embedding web page content based on an individual level learning mechanism

A computer-implemented method for enhancing fraud detection based on transactions at potentially compromised locations is provided. The method includes determining a list of potentially compromised transaction locations, storing the list of potentially compromised transaction locations, and receiving from a first transaction location a first authorization request message for a first transaction. The first authorization request message is associated with a first payment account. The method also includes determining the first transaction location based on the first authorization request message, determining if the first transaction location is in the list of potentially compromised transaction locations, and if the determination is that the first transaction location is included in the list of potentially compromised transaction locations, flagging the first payment account as potentially compromised so that a fraud score for a future transaction is associated with the first payment account is adjusted based on the potentially compromised flag.

Systems and methods for enhanced fraud detection based on transactions at potentially compromised locations

An apparatus for generating merchant clusters based upon monitored consumer transaction activity can include a processor configured to communicate with a consumer transaction database. The processor can be configured to obtain transaction activity of consumers. The processor can be configured to generate a reference cluster based on the obtained transaction activity. The processor can be configured to generate at least one regional merchant cluster for each of a plurality of different geographic regions based on the obtained transaction activity. The processor can be configured to generate a cluster labeling scheme, for each of the different geographic regions, based on comparing each regional merchant cluster for that corresponding geographic region against the reference cluster. The processor can be configured to realign cluster labels for each regional merchant cluster for each of the different geographic regions based on the corresponding cluster labeling scheme generated for the corresponding geographic region.

Methods, systems, networks, and media for generating location based clusters of merchants based on consumer transaction activity

Individual level learning mechanism

Disclosed are exemplary embodiments of systems and methods for identifying a merchant location based on transaction records for the merchant. In an exemplary embodiment, a method generally includes accessing multiple router records. Each router record is assigned to one of multiple cells of a grid structure indicative of geographic location. Each router record is associated with a target merchant. The method further includes, for each of the multiple router records, searching for a match between the router record and at least one of multiple transactions at the target merchant, based on time, and reporting one or more of the multiple cells as a location associated with the target merchant, when a number of matching router records assigned to said one or more of the multiple cells exceeds a predetermined threshold.

Systems and methods for identifying merchant locations based on transaction records

A computer-implemented method for recommending items for future purchase by a consumer based on the consumer's historical actions is provided. The method is implemented using an adaptive recommendation ('AR') computer device in communication with a memory. The method includes receiving from a recommender server computing device at least one preference vector, storing the at least one preference vector, receiving at least one consumer action including a time aspect, from a candidate consumer, and determining at least one personalized vector based on the at least one preference vector, the at least one consumer action, and the time aspect associated with each of the at least one consumer action. The at least one personalized vector represents at least one future purchase that the candidate consumer will likely conduct. The method also includes displaying at least one recommendation to the candidate consumer based on the at least one personalized vector.

Adaptive recommendation system and methods

A computer-based method for predicting consumer behavior is provided. The method is performed using a computer system coupled to a database. The method includes recording consumer data in the database for each consumer of a global population of consumers including historical purchases made by each consumer using a transaction card, defining a life event by assigning spending variables to the life event, determining a sample group of consumers that are experiencing the life event based on the consumer data stored within the database with respect to the spending variables, generating a predictive model based on historical purchases made by consumers within the sample group, and applying the predictive model to predict each consumer within the global population that will experience the life event. The predictive model is applied using the computer system. A list of consumers predicted to experience the life event within a predetermined time period is output.

Methods and systems for predicting consumer behavior from transaction card purchases

A method and system for detecting fraud in a payment card network using a pattern of transaction ticket size are provided. The method including receiving transaction information, for a current financial transaction, from at least one of a merchant point of sale (POS) device and a merchant website, the transaction information including a current transaction amount, the transaction information associated with a single payment card cardholder, retrieving a predetermined number of historical transactions for the single cardholder based on the transaction information, and generating a historical spend ticket size pattern based on average ticket size and dispersions for at least one of the same store, similar stores, and relevant merchant categories. The method further including comparing the current transaction amount to the historical spend ticket size pattern and generating a recommendation for approval or decline of the current financial transaction based on the comparison.

Systems and methods for fraud detection by transaction ticket size pattern

Systems and methods are provided for storing, sorting, and analyzing account transactions in a payment network, and for real-time data analytics. One embodiment of a real-time data storage and analytics system includes a payment network that collects and processes account transactions. The system also includes a first data cluster that includes a first data structure. The first data structure stores a set of the account transactions. Additionally, the system includes a second data cluster that in turn includes a second data structure. The second data structure stores account transactions for new account and account transactions for which there is insufficient space in the first data structure. Moreover, the system includes an application engine that generates a score for an account based on transaction data associated with the account transactions for the account.

Real-time data storage and analytics system

Exemplary embodiments of systems and methods are provided for determining detailed locations of entities. One exemplary method includes receiving at least one signal record for a communication device associated with a user. The at least one signal record includes multiple signal strengths, a temporal indicator, and an identifier unique to the communication device. The exemplary method further includes accessing multiple discrete event records associated with the entity, correlating one of the multiple discrete event records to the at least one signal record based on the temporal indicator included in the at least one signal record, and generating a location range associated with the entity, based on the signal strengths included in the at least one signal record, whereby a location of the entity, distinct from one or more other entities, is indicated by the location range.

Systems and methods for use in determining detailed locations for certain entities

According to some embodiments, an acquiring party prediction platform may receive transaction information about a plurality of transactions associated with a merchant, and different transactions may be associated with different acquiring parties. The transaction information may be analyzed to generate a prediction that a particular acquiring party may be associated with a transaction volume below a threshold volume in the future. An indication associated with the prediction may then be transmitted.

Systems and methods for predicting a merchant's change of acquirer

A method for processing payment transactions with merchant-specific fraud detection includes: storing a merchant profile, the profile including data related to a merchant including a merchant identifier and a plurality of transaction data entries, each entry including data related to a payment transaction involving the merchant including transaction data; receiving an authorization request for a payment transaction, the request including the merchant identifier and transaction data; calculating a fraud score for the payment transaction based on application of one or more scoring algorithms to the transaction data included in the authorization request and based on the transaction data included in transaction data entries in the merchant profile; transmitting the transaction data included in the received authorization request and the calculated fraud score to the merchant; and receiving a notification from the merchant indicating acceptance of risk for the payment transaction.

Growing community of inventors

Norwalk, CT, United States of America

Po Hu