US 12,339,817 B2
Methods and systems for identifying and correcting anomalies in a data environment
Irina Niyazov, Commack, NY (US); Michael Bender, Rye Brook, NY (US); and Manoj Acharya, Charlotte, NC (US)
Assigned to Charter Communications Operating, LLC, St. Louis, MO (US)
Filed by Charter Communications Operating, LLC, St. Louis, MO (US)
Filed on Aug. 30, 2022, as Appl. No. 17/899,235.
Prior Publication US 2024/0070130 A1, Feb. 29, 2024
Int. Cl. G06F 17/00 (2019.01); G06F 16/215 (2019.01); G06F 16/23 (2019.01); G06F 16/2458 (2019.01)
CPC G06F 16/215 (2019.01) [G06F 16/2365 (2019.01); G06F 16/2462 (2019.01)] 19 Claims
OG exemplary drawing
 
1. A method of detecting and responding to data anomalies, comprising:
creating a corpus of data based on architecture or standards documents;
generating classifier models based on the corpus of data;
collecting information from one or more data sources;
generating feature vectors based on the collected information;
applying the feature vectors to the classifier models to generate an analysis result;
identifying a data anomaly based on the generated analysis result;
determining a cluster that includes a significance score that exceeds a threshold value;
identifying a data consumer or data producer that has a registered interest in the data;
alerting the identified data consumer or data producer of the identified data anomaly;
receiving feedback in response to alerting the identified data consumer or data producer of the identified data anomaly;
updating the corpus of data based on the received feedback; and
performing autonomous learning of critical metrics by periodically:
evaluating a behavior of a critical metric;
classifying the critical metric based on a result of the evaluation; and
correlating the critical metric with a model from a library of model types for different signal types.