US 12,380,459 B2
Method and system for driving zero time to insight and nudge based action in data-driven decision making
Vivek Saxena, Gurgaon (IN); Kathryn Stein, New York, NY (US); Lavi Sharma, Gurgaon (IN); and Vikram Jha, Sealdah (IN)
Assigned to Genpact USA, Inc., New York, NY (US)
Filed by Genpact USA, Inc., New York, NY (US)
Filed on May 20, 2022, as Appl. No. 17/749,658.
Prior Publication US 2023/0419346 A1, Dec. 28, 2023
Int. Cl. G06Q 30/0203 (2023.01); G06Q 30/0201 (2023.01); G06Q 30/0204 (2023.01); G06N 20/00 (2019.01)
CPC G06Q 30/0203 (2013.01) [G06Q 30/0204 (2013.01); G06Q 30/0206 (2013.01); G06N 20/00 (2019.01)] 18 Claims
OG exemplary drawing
 
17. A computer-implemented method for driving zero time-to-insight and effectiveness of insight-to-nudge in a decision-making process, the method comprising:
identifying a scope associated with a data-to-action loop in the decision-making process;
determining, by a zero time-to-insight engine, a zero time-to-insight quotient for a data-to-insight loop included in the data-to-action loop;
determining, by an insight-to-nudge engine, an insight-to-nudge quotient for an insight-to-action loop included in the data-to-action loop;
determining, by a predictive model factor component, a data-to-action prediction model factor (D2A PMF) for the data-to-action loop, the D2A PMF quantifying an incremental zero time-to-insight potential for the data-to-action loop and corresponding attributes, wherein the predictive model factor component comprises a predictive machine learning model trained by:
running simulations on a plurality of data-to-action cycles for various permutations of data-to-action cycle attributes;
dividing the simulations and data-to-action cycles in real-world scenarios into a training set and a testing set; and
training and testing the predictive machine learning model based on the training set and the testing set
wherein training the predictive machine learning model includes one or more additional training steps when differences between outputs from the testing of the predictive machine learning model and labeled outputs exceed a specified threshold;
generating a nudge quotient for the data-to-action loop based on the zero time-to-insight quotient, the insight-to-nudge quotient, and the D2A PMF; and
dynamically adjusting a data collection process to at least improve data privacy based on the generated nudge quotient.