Abstract
The present disclosure relates to detecting and correcting misconfigured merchant configuration data, such as a channel indicator (e.g., card present or eCommerce) or a merchant category code, in a payment ecosystem. Cardholder transaction patterns may be generated for an individual merchant from transaction data recorded by a payment network. The patterns may include a distribution of distances between the merchant location and the billing addresses of transacting cardholders, and a distribution of transaction volume by hour within a day. The distributions may be binned and normalized. Baseline patterns may be generated for each configuration value group, each group including a large number of merchants sharing the same configuration value. The merchant pattern may be compared with the baseline pattern of the merchant's configured value using a similarity metric and a threshold. If the patterns mismatch, the merchant pattern may be compared with baseline patterns of other value groups, and if a match is found, a corrected configuration value may be proposed. Data of the matched baseline pattern may also be compared with merchant transaction data in an authorization message to detect manipulation of authorization data by a merchant or by a payment service provider or acquirer, and a correction may be proposed using the matched pattern. Proposed corrections may be provided to downstream systems such as interchange calculation and dispute handling. Deployment variants, alternative metrics, machine learning alternatives, and additional pattern features are also described.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Chen, Yuexi, "USE CARDHOLDER DISTRIBUTION PATTERN TO DETECT MERCHANT TYPE", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/12290