首页|Patent Issued for Supervised machine learning for distinguishing between risky and legitimate actions in transactions (USPTO 11875353)
Patent Issued for Supervised machine learning for distinguishing between risky and legitimate actions in transactions (USPTO 11875353)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
A patent by the inventors Abadi, Shiran (Tel-Aviv, IL), Haizler, Tamar Miriam (KfarSaba, IL), filed on September 29, 2021, was published online on January 16, 2024, according to news reporting originating from Alexandria, Virginia, by NewsRx correspondents. Patent number 11875353 is assigned to NCR Voyix Corporation (Atlanta, Georgia, United States). The following quote was obtained by the news editors from the background information supplied by the inventors: “Current loss prevention solutions treat every abnormal action in a transaction with the same severity, even if the abnormal action was merely to correct a known error associated with the transaction. “Existing solutions for cashier fraud do not distinguish well between human errors or required actions and fraudulent activities when analyzing actions that might indicated risks within a transaction. The solutions rely on simplistic rules to decide whether the actions taken during a transaction are legitimate. Such actions include, among others, voiding an item, overriding a price of an item, or returning an item. In order to give rules a more useful and tangible meaning, existing solutions often elevate a risk score with the increase in a counter value or an amount when abnormal actions are detected in a transaction.
NETL
NSTL
