US 12,406,301 B2
Early network growth warning system and method
Shiv Markam, Mandla (IN); Rupesh Kumar Sankhala, Churu (IN); Bhargav Pandillapalli, Atmakur (IN); Aniruddha Mitra, Kolkata (IN); and Akash Singh, Delhi (IN)
Assigned to MASTERCARD INTERNATIONAL INCORPORATED, Purchase, NY (US)
Filed by MASTERCARD INTERNATIONAL INCORPORATED, Purchase, NY (US)
Filed on May 20, 2022, as Appl. No. 17/749,912.
Prior Publication US 2023/0377038 A1, Nov. 23, 2023
Int. Cl. G06Q 40/04 (2012.01); G06F 18/2137 (2023.01); G06N 3/044 (2023.01); G06Q 20/40 (2012.01)
CPC G06Q 40/04 (2013.01) [G06F 18/21375 (2023.01); G06N 3/044 (2023.01); G06Q 20/4016 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A growth predictor system, comprising:
a monitor configured to receive or generate first information of a network already identified as a candidate money laundering (ML) network by an anti-money-laundering system; a prediction engine having a feed-forward neural network that operates on the first information corresponding to growth of the network already identified as a candidate ML network and being configured to predict second information indicative of a growth size of the candidate ML network at a future time, the prediction engine to predict the second information based on the first information; and a prioritization engine configured to determine a priority of the candidate ML network based on the second information and to generate a score corresponding to the priority of the candidate ML network based on one or more predetermined criteria including predicted growth size, types of accounts, identity of account owners, numbers of transactions that currently exist and are predicted to take place in the ML network, types of transaction that currently exist and are predicted to take place in the ML network, the rate of growth predicted to occur, and the type of growth predicted to occur, the prediction engine to execute one or more predictive models to generate the second information indicative of the growth size of the candidate ML network at the future time, the first information indicating one or more graphical changes that have occurred in the candidate ML network over a past period of time, wherein: the first information includes spatio-temporal data on graphs corresponding to the one or more changes that have occurred in the candidate ML network over the past period; and one of the one or more the predictive models is configured to implement an encoder decoder long short-term memory (LSTM) network to predict the growth in size of the candidate ML network based on the first information.