11 results about "Fraudulent transaction" patented technology

A system and method including: obtaining pending transactions; processing weightings for parameters associated with the pending transactions; encoding the pending transactions with the processed weightings into corresponding transaction vector representations; obtaining nearest transactions using the transaction vector representations; determining whether the pending transactions are flagged for association with at least one risky entity; in response to determining that the pending transactions are flagged for association with a risky entity, generating an alert for the nearest transactions; in response to determining that the pending transactions are not flagged, determining whether the nearest transactions are associated with at least one risky entity; in response to determining that the pending transactions or the nearest transactions are flagged for association with at least one risky entity, generating an alert for the pending transactions; and output the generated alerts to one or more computing devices via a network interface.

This will prevent fraudulent transactions by third parties using automated trading systems. [Solution] The ATM 100, which performs transactions in response to user operations, includes a transaction detection unit 220 that detects transaction operations for the user to perform a transaction, an image acquisition unit 230 that acquires an image of the user when a transaction operation is detected, an estimation unit 240 that analyzes the user's image and estimates the user's attribute information, a storage unit 210 that stores pre-registered user attribute information, a determination unit 250 that compares the estimated user attribute information with the pre-registered user attribute information and determines whether or not to perform a transaction in response to the user's operation based on the comparison result, and an output unit 260 that outputs the determination result.

A method, system, and computer program product for fraud detection receives transaction data associated with a plurality of transactions; determines, based on the transaction data, that two or more consecutive transactions associated with a same account identifier include a same value of at least one transaction parameter; in response to determining that each of the two or more consecutive transactions associated with the same account identifier includes the same value of the at least one transaction parameter, determines a difference between a value of the at least one transaction parameter associated with a first transaction of the two or more consecutive transactions and a value of the at least one transaction parameter associated with a second transaction of the two or more consecutive transactions; and determines, based on the difference, that the two or more consecutive transactions are fraudulent transactions.

The present disclosure provides a variety of solutions that enable a generative adversarial network (GAN) for fraud detection. For example, in one aspect, the present disclosure provides a system that includes a creator model and an evaluator model. The creator model is to receive real transaction data corresponding to real transactions performed across a payment network, receive fraud detection rules to identify fraudulent transactions, and generate synthetic transaction data based on the real transaction data and the fraud detection rules. The synthetic transaction data corresponds to synthetic transactions that are not actually performed across the payment network. The evaluator model is to receive the synthetic transaction data and classify at least some of the synthetic transactions as fraudulent transactions.

This disclosure relates to detecting fraudulent transactions in a multi-user account based on multimodal data that reflects the actions of multiple users of the account. A system and process may receive transaction requests associated with the multi-user account from multiple users at multiple user-interactive computers. A server receives the transaction requests from the user-interactive computers. The server may also receive multimodal data, such as audio, video, biometric, and other data that provides information about the actions of at least two of the users. The server analyzes the transaction requests and multimodal data to detect correlations between the user's actions and transaction requests that may be indicative of an unauthorized or fraudulent transaction.

A method for facilitating fraud detection based on collaborative filtering is provided. The method includes segregation of payment modes into payment mode clusters and merchants into merchant clusters based on first historical transaction data associated therewith. Further, a collaborative filtering (CF) matrix is generated based on the payment mode clusters, merchant clusters, and second historical transaction data. Additionally, CF features are created for each payment mode cluster-merchant cluster pair based on the CF matrix and second historical transaction data. A risk score machine-learning model (ML) is trained based on created CF features and non-CF features. The risk score ML model is operable to classify a transaction request to one of a fraudulent transaction request and legitimate transaction request based on the training.

The invention provides one or more consortia of networks that identify and share information about users and / or user devices interacting with the consortia. User devices may be identified, at least in part, by tag-based computer information. Computers and other devices accessing the Web carry device tags with date and time information describing when they were issued by a security tag server. A server time stamp may be inserted into time based computer tags such as a cookies indicating when they were created. Such time stamp information can be encrypted and analyzed during future attempts to access a secure network such as a customer attempting to log into an online banking account. When the time stamp information from the tag is compared to other selected information about the user, device and / or account, including but not limited to last account log-in date / time or account creation date, the invention may be used to detect suspicious activity. The invention may be use for identity-based applications such as network security, the detection of fraudulent transactions, identity theft, reputation-based communities, and law enforcement.

This invention belongs to the field of bank card fraud detection technology, and discloses a bank card fraud detection system based on the uncertainty of funds. The system includes: a user feature establishment end, a behavior analysis end, and a data update end. The user feature establishment end cleans and standardizes user data, providing a comprehensive data foundation for subsequent analysis. The behavior analysis end collects the fingerprint biometrics of the current initiator and deeply analyzes the anomalies of transaction time, region, and network environment to accurately determine fraudulent transactions. The data update end updates the user's transaction data in real time, optimizing the detection model and improving recognition accuracy. The bank card fraud detection system based on the uncertainty of funds accurately detects bank card fraud and dynamically updates user features through the quantification of fund uncertainty, biometric verification, and multi-module collaboration.

One or more devices providing improved security for any database including a smart contract or distributed application executing on a virtual machine. Embodiments provide monitoring and incident response for execution of the smart contract or distributed application running on a blockchain, or any other distributed ledger technology. An aspect of some embodiments is detection of anomalies when executing the smart contract or distributed application in real-time. Embodiments may immediately respond to the detected anomalies, e.g., by switching control to a human operator, a reactive machine, or a machine learning operator. In an embodiment, operator response may include i) freezing execution, ii) resuming execution, or iii) reverting execution. Accordingly, in the event of an erroneous or fraudulent transaction, the operator machine may revert the smart contract or distributed application to a previously valid state, thereby preventing loss of digital assets or funds allocated to the smart contract or distributed application.

Disclosed herein are system, method, and computer program product embodiments for detecting a fraudulent transaction. In an embodiment, runtime data comprising one or more metrics are received from a client device. A quantum computer is configured so as to simulate transition probabilities of one or more Markov chains by processing a plurality of qubits using a quantum circuit. A fraudulent transaction is detected based on at least a measurement value of the plurality of qubits. One or more packets configured to stop the fraudulent transaction are transmitted to the client device. A respective initial state of each qubit of the plurality of qubits is obtained by encoding the runtime data.