System and method for machine learning-based real-time electronic data quality checks in online machine learning and AI systems

A standalone data quality learning module for machine learning-based real-time electronic data quality checks, the module comprising:

• a memory device with computer-readable program code stored thereon;
• a neural network device;
• a communication device;
• and
• a processing device operatively coupled to the memory device and the communication device, wherein the processing device is configured to execute the computer-readable program code to:
• continuously monitor and analyze, in real-time, an input data stream via a data quality learning module according to a set of data quality criteria, wherein the input data stream comprises one or more transaction requests, wherein the set of data quality criteria comprises historical data gathering of data quality patterns and associated static exception conditions, real-time data gathering of data patterns, and feeds from known adversarial attack types;
• based on analyzing the input data stream, dynamically adjust a set of exception criteria;
• transmit a data quality output from the data quality learning module to a machine learning module, wherein the data quality output comprises a data quality score generated based on the set of data quality criteria;
• based on the data quality output, generate, via the machine learning module, a machine learning output, wherein the machine learning output comprises one or more recommended actions for processing the one or more transaction requests;
• and
• transmitting real-time feedback from the machine learning module to the data quality learning module, wherein the real-time feedback comprises at least one of a measure of data quality and one or more recommendations to the data quality learning module for processing the input data stream.

A computer-implemented method for machine learning-based real-time electronic data quality checks, the method comprising:

• continuously monitoring and analyzing, in real-time, an input data stream via a data quality learning module according to a set of data quality criteria, wherein the input data stream comprises one or more transaction requests, wherein the set of data quality criteria comprises historical data gathering of data quality patterns and associated static exception conditions, real-time data gathering of data patterns, and feeds from known adversarial attack types;
• based on analyzing the input data stream, dynamically adjusting a set of exception criteria;
• transmitting a data quality output from the data quality learning module to a machine learning module, wherein the data quality output comprises a data quality score generated based on the set of data quality criteria;
• based on the data quality output, generating, via the machine learning module, a machine learning output, wherein the machine learning output comprises one or more recommended actions for processing the one or more transaction requests;
• and
• transmitting real-time feedback from the machine learning module to the data quality learning module, wherein the real-time feedback comprises at least one of a measure of data quality and one or more recommendations to the data quality learning module for processing the input data stream.