Abstract: A computer-implemented method for predicting purchasing behaviour of digital wallet users for wallet-based transactions made by the digital wallet users is provided. The method includes identifying payment card users having a profile characteristic that matches the profile characteristic of the digital wallet users. The method also includes retrieving historical transaction data by querying a transaction database, wherein the historical transaction data relates to a plurality of historical transactions settled for the identified payment card users. The method further includes identifying, from the retrieved historical transaction data, transaction information of the identified payment card users to derive purchasing behaviour of the identified payment card users. The method also includes predicting the purchasing behaviour of the digital wallet users for wallet-based transactions made by the digital wallet users, based on the derived purchasing behaviour of the identified payment card users.

Abstract: Method and system for providing a housing recommendation to a user. The method comprising: determining, using a recommendation module, a spending behaviour of the user based on transaction data relating to electronic payment transactions involving the user; dividing, using the recommendation module communicatively coupled with a geographic information system (GIS), a geographical area into a plurality of localities based on geographic information from the GIS; determining, using the recommendation module, an average spending behaviour of residents in each locality based on transaction data relating to electronic payment transactions involving the residents; assigning, using the recommendation module, a base score to each locality based on a presence of residents in the locality with similar spending behaviour to the user; and transmitting, to a user output module communicatively coupled with the recommendation module, the housing recommendation that is based on the base score of each locality.

Abstract: A computer system and method for determining a cardholder location during Card Not Present (CNP) transaction. The system comprises first, second and third databases, and a location engine comprising a processor operable by instructions stored in memory that, when executed by the processor, cause the processor, during use of the payment card during a CNP transaction with a first one of the plurality of merchants determine the location of the cardholder during the CNP transaction.

Abstract: A method for estimating potential demand at a prospective merchant location for a merchant of a prospective merchant industry is provided. The method includes receiving transaction data and determining from the transaction data a first set of transactions, which include transactions carried out by consumers with consumer origin locations within an area that includes the prospective merchant location.

Abstract: Disclosed is method that includes processing the past registered interest data against the past purchase data to identify past purchases resulting from registered past interest so as to determine a period between each of the identified past purchases and the registered past interest; deriving parameters, from one or more of the determined periods, for a prediction algorithm configured to forecast the interval between the interest registered for the particular good or service and purchase of the particular good or service; processing, using the prediction algorithm, present data of interest registered on one or more goods or services that are related to the particular good or service to calculate an estimate of the interval between the registered present interest and the purchase of the particular good or service; and planning the product related activities in response to the estimated interval. Related apparatus and a non-transitory computer readable medium are also disclosed.

Abstract: A computer-implemented method of detecting associated payment devices comprises: receiving, by a processor from a payment network, transaction data representing a plurality of transaction records, a transaction amount, a transaction time, a terminal identifier, and a result code indicating success or failure of a transaction; identifying from said transaction records a declined transaction associated with a first payment device identifier, and determining a terminal identifier and transaction amount associated with the declined transaction; identifying from said transaction records a subsequent transaction occurring within a predetermined window of the declined transaction, the subsequent transaction being associated with the same terminal identifier and the same transaction amount as the declined transaction, the subsequent transaction being associated with a second payment device identifier that is different from the first payment device identifier; and storing, in a database, an association between the sec

Abstract: A computer-implemented method for computing a rating for a service provider is provided. Thee method comprising operations of (a) receiving, by a transaction analysis component, transaction data representing past transactions performed by customers with the service provider via a payment network, (b) calculating, by the transaction analysis component, a transaction-based measure characterizing the past transactions during a pre-defined period; (c) receiving, by a service provider rating component, a review score indicative of a customer rating for the service provider; and (d) computing, by the service provider rating component, the rating using the transaction-based measure and the review score, said rating being indicative of a quality of service associated with the service provider. A system for carrying out the method is also provided.

Abstract: A top-down clustering method and apparatus recursively processes clusters of documents by first extracting features from the documents comprising the cluster, then using the extracted features to generate sub-clusters and finally using the generated sub-clusters to develop topics and identifiers for each sub-cluster. This process is repeated for each cluster and sub-cluster in a recursive manner so that clustering is performed using features extracted from each document in a cluster to perform sub-clustering. Feature extraction is performed by using frequency counts of terms taken from each document in a cluster and discarding terms falling outside of predetermined boundaries computed based on the total number of documents in the cluster. After bounding, the number of tokens is reduced prior to clustering by means of a correlation technique, such as a PCA model.