US 12,443,506 B2
Real-time remote system shutdown prediction
Itay Margolin, Petah Tikva (IL); Aleksandr Kim, Tel-Aviv (IL); and Yair Horesh, Kefar-Sava (IL)
Assigned to Intuit Inc., Mountain View, CA (US)
Filed by Intuit, Inc., Mountain View, CA (US)
Filed on Aug. 30, 2023, as Appl. No. 18/240,234.
Prior Publication US 2025/0077382 A1, Mar. 6, 2025
Int. Cl. G06F 11/34 (2006.01)
CPC G06F 11/3452 (2013.01) [G06F 11/3447 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for detecting data collection errors, comprising:
receiving, from a remote system, error data;
processing the error data with a plurality of regression models to generate a plurality of predicted error rates over a plurality of time intervals, wherein:
each respective regression model of the plurality of regression models is configured to predict an error rate over a respective time interval different than each other regression model in the plurality of regression models, and
the plurality of predicted error rates includes at least one predicted error rate for each different time interval in the plurality of time intervals;
determining an error mode by applying a set of policy rules configured for determining the error mode to the plurality of predicted error rates;
in response to determining the error mode, declaring a remote system data outage; and
pausing collection of data from the remote system for an outage period based on at least one policy rule in the set of policy rules.