US 12,073,186 B1
Machine learning report generation
Lei Guang, Montreal (CA); and Lars Soldahl, Antioch, CA (US)
Assigned to Jumio Corporation, Sunnyvale, CA (US)
Filed by Jumio Corporation, Palo Alto, CA (US)
Filed on Sep. 30, 2021, as Appl. No. 17/491,446.
Int. Cl. G06F 40/40 (2020.01); G06F 40/174 (2020.01); G06F 40/186 (2020.01); G06N 20/00 (2019.01)
CPC G06F 40/40 (2020.01) [G06F 40/174 (2020.01); G06F 40/186 (2020.01); G06N 20/00 (2019.01)] 19 Claims
OG exemplary drawing
 
1. A computer implemented method comprising:
receiving, using one or more processors, a case;
mapping, using the one or more processors, a first label to a series of reportable activities in one or more jurisdictions, the first label associated with the case by a machine learning model based on the first label meeting a first threshold confidence level;
prepopulating, using the one or more processors, one or more reports, the one or more reports reporting the series of reportable activities in the one or more jurisdictions;
generating, using the one or more processors, a template-based narrative, wherein a template is based on the first label, the first label associated with a first category of activities;
generating, using the one or more processors, a natural language narrative by applying natural language generation associated with the first category of activities to the template-based narrative;
prepopulating, using the one or more processors, a form field with the natural language narrative;
receiving, based on user input, a modification to an initial set of labels associated with the case, the modification including the first label; and
retraining the machine learning model that generates initial sets of labels associated with cases.