CPC G06Q 20/407 (2013.01) [G06N 20/00 (2019.01); G06Q 20/4016 (2013.01); G06Q 20/405 (2013.01)] | 20 Claims |
1. A computer-implemented method comprising:
receiving, from a client device, a dispute request comprising information associated with an initial disputed transaction within a user account corresponding to the client device;
determining a dispute classification for the initial disputed transaction by analyzing, utilizing a combination of machine learning models and heuristic models, metadata of the initial disputed transaction, the dispute classification corresponding to a dispute classification category, the dispute classification category comprising one of: an unauthorized transaction, a device malfunction, a non-receipt of goods, or a non-matching description of goods received;
identifying one or more potential transactions within the user account to generate a multi-transaction dispute package comprising the initial disputed transaction and the one or more potential transactions by:
selecting, based on the dispute classification category, a machine learning model from a plurality of machine learning models;
encoding the dispute classification; and
utilizing the selected machine learning model to process the encoding of the dispute classification and output the one or more potential transactions;
providing, for display on the client device, the one or more potential transactions; and
based on receiving, from the client device, an indication of a selection of at least one additional disputed transaction from the one or more potential transactions, generating the multi-transaction dispute package comprising the initial disputed transaction and the at least one additional disputed transaction.
|