Abstract: Training a machine learning model includes selecting a subset of training transactions from a plurality of training transactions to be used for classifying transactions as either fraudulent or genuine, and classifying transactions as either fraudulent or genuine. Additionally, selecting the subset of training transactions includes clustering the plurality of training transactions into a plurality of clusters based on a similarity measure. Each training transaction includes an indication whether the training transaction is fraudulent or genuine. Further, selecting the subset of training transactions includes selecting a first proportion of the training transactions of a cluster of training transactions for the subset of training transactions if the cluster comprises at least one fraudulent training transaction, and selecting a second proportion of the training transactions of a cluster of training transactions for the subset of training transactions if the cluster is free of fraudulent training transactions.

Abstract: A fraud prevention server and method. The fraud prevention server includes an electronic processor and a memory. The memory includes an online application origination (OAO) service. The electronic processor is configured to perform feature encoding on one or more categorical variables in a first matrix to generate one or more feature encoded categorical variables, generate a second matrix including the one or more feature encoded categorical variables, generate a feature encoded OAO model by training an OAO model with the second matrix, receive information regarding a submission of an online application on a device, determine a fraud score of the online application based on the information that is received and the feature encoded OAO model, the feature encoded OAO model being more precise than the OAO model, and control the client server to approve, hold, or deny the online application based on the fraud score that is determined.

Abstract: Techniques are described herein for pattern identification using enhanced event stream processing. In an embodiment, an event instance is received via a programming interface, the event instance representing an occurrence of a corresponding type of real-world event and comprising one or more variable values pertaining to the occurrence. A ruleset is periodically evaluated as applied to the event instance including, for each of one or more reevaluations of the ruleset as applied the event instance, binding one or more variables of one or more rules of the ruleset to one or more variable values of the event instance and binding one or more other variables of one or more rules of the ruleset to one or more just-in-time variable values, the one or more just-in-time variable values received via one or more programming interfaces in context of the reevaluation.

Abstract: Techniques are disclosed in which a computer system generates a transaction network graph from an initial set of transactions including known labels and attributes. The computer system may generate first and second matrices using first and second graph embedding routines from a training set of transactions that includes a first subset of transactions in the network graph. The first routine is based on anomalies in related transactions occurring at nodes in the transaction network graph that are multiple hops away while the second routine is based on anomalies in neighborhoods of similar transactions. In some embodiments, the computer system generates a final embedded matrix from the first and second matrices and uses the final matrix and a testing set of transactions that includes a second subset of transactions in the graph to train a machine learning model, where the trained model usable to determine whether unlabeled transactions are anomalous.

Abstract: Enhanced authentication at resource event apparatuses of a card device and, in some instances, the user of the card device. Card device-specific Non-Fungible Tokens (NFTs) are generated and, in some instance, user credential-specific NFTs are generated. The NFTs are stored within a distributed ledger of a distributed trust computing network, which provides verification as the uniqueness and authenticity of the NFTs. Once a resource event is initiated at a resource event apparatus, card device-related information and, in some instance, user credential-related information is received, and the information is compared to the corresponding NFT as a means of authenticating the card device and, in some instances, the card device and the linked user of the card device. As a result, the use of imitation card devices is prevented and, in some instances, the identity of the user of the card device is authenticated and deemed to be linked to the card device.

Abstract: In a method for broadcasting account provisioning information, user datum encryption information is transmitted to a plurality of account processing systems. A request to share account provisioning information is received from a user device associated with an account holder. A user datum of the account holder is encrypted using the encryption information and transmitted to the account processing systems. A response is received from at least one of the account processing systems, the response including an indication that the account holder has a transaction account processed by that account processing system. A request for confirmation that the account provisioning information should be shared is sent to the user device and a confirmation response is received from the user device that includes permission to share account provisioning information. Account provisioning information is then transmitted to one or more of the account processing systems.

Abstract: Systems and methods for fraud monitoring are disclosed, including: receiving a transaction request associated with a first instrument of a user; extracting, characteristics of the transaction request; identifying, by the first processor, user data based on the transaction request; determining a fraud severity value and notification value based on inputting the characteristics and user data into a fraud machine learning model; performing a first fraud action based on the fraud severity value; wherein the first fraud action is at least one selected from the group of locking the first instrument for a period of time, deactivating the first instrument, and electronically transmitting a first query message to a user device associated with the first instrument; and transmitting a fraud notification based on the notification value, wherein the fraud notification includes severity information associated with the fraud severity value.

Abstract: Methods, systems, devices, and computer-readable media for a payment delegation and linking system are provided. The system may generate a delegation request to send to an individual for payment of a portion of a value of a transaction. The delegation request may be sent to the individual with a payment URL embedded with a token. Upon initiation of payment in response to the delegation request, the token may be embedded in an unused field of a payment transaction record generated at the individual's device and may be transmitted from the individual's device to a payment processing system for processing and settlement. The payment transaction record may ultimately be received at the requestor's bank with the token intact. The token may be used to link the received payment transaction with the original delegated transaction. The original delegated transaction and the received payment transaction may be displayed together in a transaction summary.

Abstract: Disclosed is a system, apparatus, and business method for using financial instruments to provide investment incentive to fund disaster mitigation projects and/or ecosystem restoration projects triggered by real world events detected at set points by sensors and analyzed by computing devices. The detection of events results in the computational analysis and determination of a characteristics of a resilience bond or ecosystem-service bond to be issued or returned to investors. Examples involve sensors used to detect events, including flood damage mitigation, earthquake damage mitigation, or air quality improvement. In one preferred embodiment of the disclosure an investor may make an initial payment into a financial instrument being a resilience bond used to fund a mitigation project, said financial instrument pays phases to the mitigation project developer at the completion of phases of the project that occur prior to the occurrence of a disaster.

Abstract: Systems and methods can include a computing system configured to receive transaction information corresponding to a transaction between a merchant and a customer. The computing system can be configured to receive one or more images captured via a camera during the transaction. The computing system can be configured to analyze the one or more images captured by the camera to detect one or more features relating to the transaction in the one or more images. The computing system can be configured to generate a designation for the transaction based on the one or more features detected in the one or more images. The designation can comprise at least one of verified or unverified. The computing system can be configured to store the transaction information and the one or more images with the designation for the transaction in an account associated with the customer maintained in an accounts database.

Abstract: A method and system for providing secure data communication between two or more user terminals in a communication network is proposed. The method may include receiving, from a first user terminal, a first instruction set configured to request that a private account be converted into a group account, the private account owned by a user of the first user terminal. The method may also include generating the group account jointly owned by at least the user of the first user terminal and a user of a second user terminal. The method may also include controlling a first database to store a second information set relating to the group account, and generating a second virtual database different from the first database. The method may also include copying the second information set stored in the first database and recording the copied second information set in the second virtual database.

Abstract: A user terminal including a graphical user interface (GUI) for providing secure data communication with a server in a communication network is proposed. The user terminal may include a GUI configured to receive one or more user inputs and display data received from the server. The GUI may include a first screen configured to display a first instruction set for converting a private account to a group account and a second screen configured to display a group account information set received from the server. The user terminal may cause the server to generate the group account, control a first database of the server to store a second information set relating to the group account, generate a second virtual database of the server different from the first database, and copy the second information set stored in the first database and record the copied second information set in the second virtual database.

Abstract: Aspects of the disclosure relate to a smart contactless card to detect real-time suspicious card readers or other fraudulent devices. Prior to a transaction, the smart contactless card detects suspicious card readers or fraudulent devices. An alert may be generated upon detection of any suspicious or fraudulent card reader. In some arrangements, the smart contactless card may utilize machine learning models or machine learning capabilities to detect suspicious card readers. The smart contactless card may pair with other smart contactless cards to detect and alert users to suspicious card readers or other fraudulent devices. The paired smart contactless cards may share information regarding suspicious card readers or fraudulent devices over a semi-autonomous data-sharing network. A vulnerability score may be generated and used to determine if a card reader or other payment device is suspicious.

Abstract: An apparatus, method, and non-transitory medium for betting on weather predictions are disclosed herein. A plurality of longitude and latitude coordinate pairs may be received from a remote device. The display device may render data representing a binary option based on whether a storm will make landfall within the certain radius. A request to purchase the binary option may be received. It may be determined whether the landfall location corresponds to a zip code associated with the given longitude latitude pair in the memory.

Abstract: The present application discloses a payment information processing method, apparatus, device, and computer readable storage medium. The payment information processing method includes: obtaining an environment image collected by a mobile acquiring device, in response to an acquiring request sent by the mobile acquiring device; determining a payment environment confidence level of the mobile acquiring device according to the environment image, under a condition that the mobile acquiring device obtains payment card information; determining a payment information processing result according to the payment environment confidence level.

Abstract: Systems and methods are disclosed herein for improving machine learning of a data set. In one example, the method may include training a predictive model on an initial data set comprising labeled data, wherein the training is performed in an active learning system. The method may further include generating a set of parameters based on the training and introducing an unlabeled data set into the predictive model. According to some embodiments, the method may further include applying the set of parameters to the unlabeled data set, generating a set of predictions associated with the applied set of parameters and calculating a first uncertainty score and a second uncertainty score associated with the generated set of predictions. Moreover, the method may also include modifying the data set based on the first uncertainty score, and modifying the predictive model based on the second uncertainty score.

Abstract: A method for optimizing refunds for suspected or detected fraudulent transactions includes receiving a chargeback analysis request for a potential chargeback transaction from a merchant or a payment processor, extracting identifying information of transactions associated with the chargeback transaction from the chargeback analysis request, searching for a chargeback analysis profile in a profile database, determining whether the chargeback analysis profile exists in the profile database, upon determining that the chargeback analysis profile does not exist in the profile database, obtaining a new fraud analysis profile, determining, based on the chargeback analysis profile, a first probability that the potential chargeback transaction will result in a chargeback, determining, based on the chargeback analysis profile, a second probability that the potential chargeback transaction will result in a chargeback after a proactive electronic refund transaction, and generating a proactive electronic refund transaction

Abstract: Exemplary embodiments of the present disclosure directed towards an artificial intelligence system and method for performing settlements based on activity, user groups, and set of virtual/physical entities. The artificial intelligence system comprises artificial intelligence settlement engine configured to enable users for activity and provide activity templates to the artificial intelligence settlement engine from the computing device by the users. Customize the activity templates fully/partially based on dynamic needs and then provide the customized activity templates to the artificial intelligence settlement engine. The artificial settlement engine configured to integrate settlement process steps and/or rules into the settlement process sequence on the artificial intelligence settlement engine and then map settlement process steps and/or rules with the activity templates by the artificial intelligence settlement engine.

Abstract: A method for authenticating a counterparty to a digital transaction includes obtaining, at a mobile terminal from a unverified counterparty, characteristic content associated with the digital transaction to be displayed in a trusted user interface provided by the mobile terminal, sending, by the mobile terminal, data associated with the characteristic content to an authentication server; and obtaining, from the authentication server, a result of an authentication judgment by the authentication server, the authentication judgment based on the data associated with the characteristic content and an item of reference content.

Abstract: Aspects of the disclosure relate to monitoring an automatic teller machine (ATM) network and determining anomalous fault behavior in the ATM network. A computing device may determine historical fault volumes in the ATM network and generate a time-series model of the fault volumes. The computing platform may predict future fault volumes based on the generated time-series model. Based on the predicted future fault volumes and actual future fault volumes, the computing platform may determine anomalous fault behavior. The time-series model may be based on exogenous factors associated with ATM network operations.