Abstract: A method includes extracting, by an analysis computer, a dataset including initial vector representations for each of a plurality of user nodes and for each of a plurality of resource provider nodes. The analysis computer can then generate updated vector representations as new interaction data arrives over time, and use them to perform predictions of future interactions.

Abstract: A computer-implemented method includes: generating a plurality of candidate groups based on transaction data associated with each user of a plurality of users, each candidate group of the plurality of candidate groups including a subset of users that conducted transactions with a common merchant within a threshold time; generating a plurality of actual groups based on the plurality of candidate groups and the transaction data associated with each user of the plurality of users, each actual group of the plurality of actual groups including a subset of users grouped together in at least a threshold plurality of candidate groups; and for an actual group of the plurality of actual groups, determining a classification associated with a merchant for the actual group based on transaction data for transactions conducted between the merchant and each user of the actual group.