US 12,261,872 B2
Enhanced machine learning refinement and alert generation system
Corentin Petit, Paris (FR); Jacob Albertson, London (GB); Marissa Kimball, London (GB); Paul Baseotto, Poole (GB); Pierre Cholet, Berlin (DE); Timur Iskhakov, London (GB); and Victoria Galano, Paris (FR)
Assigned to Palantir Technologies Inc., Denver, CO (US)
Filed by Palantir Technologies Inc., Denver, CO (US)
Filed on Aug. 16, 2021, as Appl. No. 17/445,172.
Claims priority of provisional application 63/199,280, filed on Dec. 17, 2020.
Prior Publication US 2022/0201030 A1, Jun. 23, 2022
Int. Cl. H04L 9/40 (2022.01); G06F 9/451 (2018.01); G06N 20/00 (2019.01)
CPC H04L 63/1433 (2013.01) [G06F 9/451 (2018.02); G06N 20/00 (2019.01); H04L 63/1425 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
by a system of one or more computers,
accessing a plurality of datasets storing customer information comprising, at least, a plurality of transactions associated with a plurality of customers, each transaction indicating a plurality of features;
generating individual risk scores for the plurality of customers based on the customer information, wherein generating the risk scores comprises:
identifying occurrences of scenario definitions, wherein each scenario definition specifies one or more object types of a plurality of object types and expressions utilizing the object types and one or more of the features,
wherein a data pipeline is applied to the datasets, wherein the data pipeline causes extraction of object types in accordance with an ontology,
wherein an occurrence of a scenario definition indicates satisfaction of the specified expression with respect to customer information, and
providing the identified occurrences and customer information as input to one or more machine learning models, wherein the machine learning models assign respective risk scores to the customers; and
causing presentation, via a user device, of an interactive user interface, wherein the interactive user interface:
presents summary information associated with the risk scores, wherein the interactive user interfaces enables an investigation into whether a particular customer is exhibiting risky behavior, and
responds to user input indicating feedback usable to update the one or more machine learning models or scenario definitions,
wherein the feedback triggers updating of the machine learning models.