Ncr Corporation

Tali Shpigel

Itamar David Laserson

Avishay Farbstein

Loran Halfon

Mor Zimerman Nusem

A machine-learning algorithm is trained with features relevant to transaction exceptions, distributions of items in transaction mapped to product hierarchies, and operator data. The trained algorithm is trained to predict whether a given transaction requires a transaction exception for potential fraud or for management approval. The trained algorithm is then provided a set of in-progress input data for an in-progress transaction being processed on a transaction terminal. Output from the trained algorithm is used to determine whether the in-progress transaction is allowed to continue processing unabated or whether the in-progress transaction is to be suspended with a transaction exception requiring a manager override or security credential to continue processing.

Transaction exception and fraud processing

A transaction is identified for a partial rescan security check based at least in part on basket items of the transaction. A total number of rescan items from the basket items is identified for rescan; the total number of rescan items selected for rescan is less than a total number of the basket items in the transaction. Based on the basket items and transaction features for the transaction, item categories or item departments are identified from which the total number of rescan items are to be selected from the basket items for the rescan security check. The total number of rescan items and the item categories for selection are provided to an attendant terminal for the rescan security check. The rescan security check is processed to determine whether the transaction was associated with theft or not associated with theft.

Data-driven partial rescan precision booster

A machine-learning algorithm is trained with features relevant to basket data for items of transactions. The trained algorithm is trained to predict whether a given transaction is more or less likely to be associated with theft being engaged in by a transaction operator for the transaction. The trained algorithm is then provided basket data for a given transaction and produces as output a theft prediction value. When the theft prediction value exceeds a configured threshold value, the transaction is flagged for manual intervention or the transaction is flagged for subsequent manual verification.

Data-driven machine-learning theft detection

Various embodiments herein each include at least one of systems, methods, and software for model-based data validation to identify when self-scan checkout data requires validation. Some embodiments, in the form of a method includes receiving, via a network from a self-scanning device, a self-scan dataset of items for purchase within a purchase data processing transaction and evaluating the self-scan dataset to determine whether to require a rescan of items represented in the self-scan dataset. In such embodiments when a rescan is determined to be required, the method includes transmitting via the network to at least one of the self-scan device and at least one device of a store employee data indicating a rescan is required. However, when a rescan is not determined to be required, the method includes permitting the purchase data processing transaction to proceed.

Growing community of inventors

Bnei Brak, Israel

Tali Shpigel