US 12,141,818 B2
Systems and methods for linking indices associated with environmental impact determinations for transactions
Sangam Verma, Gurugram (IN); Rohit Chauhan, Somers, NY (US); Athanasia Xeros, St. Louis, MO (US); Karamjit Singh, Gurugram (IN); Nitendra Rajput, Gurgaon (IN); Tanmoy Bhowmik, Bangalore (IN); and Aniruddha Mitra, Gurgaon (IN)
Assigned to MASTERCARD INTERNATIONAL INCORPORATED, Purchase, NY (US)
Filed by MASTERCARD INTERNATIONAL INCORPORATED, Purchase, NY (US)
Filed on Oct. 5, 2021, as Appl. No. 17/494,768.
Claims priority of provisional application 63/088,438, filed on Oct. 6, 2020.
Prior Publication US 2022/0108328 A1, Apr. 7, 2022
Int. Cl. G06Q 30/018 (2023.01); G06N 3/04 (2023.01); G06Q 10/063 (2023.01)
CPC G06Q 30/018 (2013.01) [G06N 3/04 (2013.01); G06Q 10/063 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A computer-implemented method for determining an environmental impact associated with one or more transactions, the method comprising:
accessing transaction data representative of a plurality of transactions, each of the transactions involving a user and one of multiple merchants, each of the merchants being associated with a merchant category code;
accessing an environmental impact index indicative of environmental impact of a plurality of merchants, the environmental impact index including an environmental impact value for each of a plurality of categories, the plurality of categories of the environmental impact index being independent of the merchant category codes;
generating a graph based on the users and merchants included in the transaction data, the graph including a node for each one of the users, a node for each one of the merchants, and edges between the nodes whereby the graph is representative of the plurality of transactions;
determining a mapping between ones of the merchants involved in the transaction data and the environmental impact indicated by the environmental impact index, based on the graph and the accessed environmental impact index by:
iteratively applying a graph convolution network (GCN) to generate a vector representing environmental impact of the ones of the merchants;
learning, via a graph neural network (GNN), embeddings for the graph, for each node representative of the ones of the merchant, which include, for each node: features of the node, features of neighboring edges of the node, and states of neighboring one(s) of the nodes for said node, the features of the node including merchant variables of the merchant represented by said node;
stacking the embeddings, along with output vectors for each node, into a matrix representing environmental impact of the ones of the merchants represented by the nodes; and
combining, via regression and weighted learning, the vector from the GCN and the matrix from the GNN to thereby provide the mapping; and
publishing the mapping between the ones of the merchants and the environmental impact indicated by the environmental impact index, to thereby enable the user to reduce environmental impact associated with subsequent transactions.