Patent · US Active

Artificial intelligence based methods and systems for predicting merchant level health intelligence

US12198187B2 · kind B2 · utility

0Cited by
3References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateMar 16, 2022
Grant dateJan 14, 2025
Priority date
Expiry dateJun 1, 2042

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N7/01
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Embodiments provide methods and systems for predicting merchant level health intelligence. Method performed by server system includes accessing, from a transaction database, payment transaction data of a merchant for a period of time. The payment transaction data includes information of payment transactions between a plurality of cardholders and the merchant for the period of time. The method includes aggregating merchant transaction attributes based, at least in part, on the payment transaction data and calculating a merchant membership vector based, at least in part, on the merchant transaction attributes by applying a soft-clustering model over the merchant transaction attributes. The method further includes predicting a commercial credit score and a merchant delinquency rate associated with the merchant based, at least in part, on the merchant membership vector. The commercial credit score and the merchant delinquency rate are predicted based, at least in part, on a multi-task learning model.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.