Risky actions versus non-risky actions in a transaction are identified and a fraud score associated with probabilities of the risky actions is updated accordingly for purposes of determining whether the transaction is likely or not likely to be associated with fraud. A machine-learning model is trained to predict the risky actions versus non-risky actions of a transaction based on the transaction features as a whole and compare the predicted action labels of risky and non-risky versus the actual actions taken in the transaction to calculate probabilities of risky actions taken and output a risk or fraud score based thereon. Higher probabilities correlate with lower risk scores and vice versus.

Systems and methods are disclosed for payment transactions, alerts, dispute settlement, and settlement payments, using multiple blockchains. One method includes: entering, in a first blockchain, a transaction identifier indicating the initiation of and identification of a transaction; receiving an identifier of a currency or cryptocurrency account for participants of the payment transaction; performing one or more iterations of: identifying a new transaction event in the series of transaction events stored in the first blockchain; presenting the new transaction event to participants of the transaction, wherein the presentation enables a participant to indicate a dispute of an attribute of the transaction; relaying one or more attributes of the transaction to a second blockchain for processing the transaction; and receiving, from the second blockchain, an indication of a transfer of funds between the two or more participants, using the identifiers of the currency or cryptocurrency accounts of the two or more participants.

Examples presented herein describe secure point of sale (POS) operations. One example is a method including receiving, at a mobile device, a communication associated with a custom tender payment for a secure transaction payment, where the mobile device includes a custom payment application configured for the secure transaction payment with a point of sale (POS) device. The method includes receiving account data associated with a user account selection for the custom tender payment for the secure transaction payment, transmitting an account signal including the account data, where when the account signal is transmitted to an integration server as part of the secure transaction payment, the account signal is not transmitted to the POS device, and receiving an authorization communication, where when the authorization communication is received by the mobile device from the integration server, the authorization communication is not communicated to the mobile device via the POS device.

There is provided a computing system for optimizing a plurality of fraud detection strategies used to generate a corresponding set of potentially fraudulent transactions. The system determines an overall fraud value such as an average fraud value for each transaction based on pre-defined factors and identifies a particular strategy having a highest average fraud value for its fraudulent transactions as a highest priority on a ranked list of strategies. The system is configured to remove each transaction from the remaining other strategies if the same as the fraudulent transactions in the identified strategy and calculate an average fraud value for the remaining other strategies. The system then ranks the next highest priority fraud detection strategy having the highest average fraud value while removing its corresponding transactions flagged from other remaining strategies and repeat the ranking until all the strategies have been ranked and apply the ranked list to subsequent transactions.

Methods, systems, and computer program products for learning continuous embedding space of real time payment (RTP) transactions are provided. A method may include receiving RTP data including a plurality of attributes, including a sender and a receiver. One attribute is selected as a target attribute. The remaining attributes are input into a first machine learning model (e.g., NLP model), comprising at least one embedding layer and one hidden layer, which is trained to predict the target attribute. After the model is trained, each of the remaining attributes are converted to a first vector using the at least one embedding layer of the machine learning model to form a first set of vectors. The first set of vectors are stored and subsequently input into a second machine learning model to perform at least one second task different than the first task.

Embodiments provide methods and systems for identifying a re-routed transaction. Method performed by processor includes retrieving a plurality of transaction windows from a transaction database. Each transaction window includes a transaction declined under a restricted MCC. The method includes accessing a plurality of features associated with each transaction of each transaction window from the transaction database. The method includes predicting an output dataset of a plurality of reconstructed transaction windows based on feeding the input dataset to a trained neural network model. The method includes computing a corresponding reconstruction loss value for each transaction of each transaction window. The method includes comparing the corresponding reconstruction loss value for each transaction with a pre-determined threshold value. The method includes identifying the transaction as a re-routed transaction corresponding to the transaction declined under the restricted MCC, if a corresponding reconstruction loss value for a transaction is higher than the pre-determined threshold value.

Exemplary embodiments are disclosed of systems and methods for monitoring for and lowering the risk of addiction-related or restriction violation-related behavior(s).

Information security processes, systems, and machines for authenticating users, wirelessly detecting a user's local devices, calculating a trust score based on the local devices, and setting a transaction limit are disclosed. An ATM or POS machine can read a card, authenticate a user, and wirelessly read MAC or other unique hardware addresses for one or more of the user's local devices. Trust scores can be calculated based on the number of local devices that are detected in relation to the number of the user's devices that are registered, the historical presence of the user's devices during prior transactions, historical usage of the ATM or POS machine, geolocating, biometric authentication(s), etc. Dynamic transaction limits, types, and rights may be set for transactions corresponding to the trust score values. Transactions may be conducted wholly or partially in a contactless fashion.

A method includes receiving a payment request that indicates a card identifier corresponding to a payment instrument to be used for payment. The method further includes determining that the card identifiers fails to satisfy at least one card activity criteria. Additionally, method includes subsequent to determining that no failed authorization attempts were performed for the card identifier within a previous time period, retrieving, from a database, a decline probability score associated with the card identifier. The method also includes based on the decline probability score, determining whether to transmit an authorization request for the card identifier prior to processing the payment request.

The present disclosure relates to a system and computer-implemented method for fulfilling an order request. The method includes receiving information regarding a transaction between a user and a merchant, and information regarding an error while processing the transaction. Further, based on a received assurance value from the merchant, requesting one or more entities for providing at least one of a partial or total of the assurance value. Furthermore, in response to the request, a contribution value is received from each of the one or more entities, where the contribution value is one of the partial or total of the assurance value. Finally, an assurance message is provided to the merchant indicating successful payment for fulfilling the order request of the user upon determining the contribution value received from the one or more entities is summing up to the assurance value.