| CPC G06F 16/215 (2019.01) [G06F 3/04847 (2013.01)] | 12 Claims |

|
1. A computer-implemented method comprising:
at a record resolution service being executed by one or more computer devices:
receiving an unmapped data record comprising a plurality of data fields;
generating a resolution candidate record for the unmapped data record, wherein generating the resolution candidate record includes:
computing, via a record correlator machine learning model, a plurality of record correlation inferences between the unmapped data record and each of the plurality of validated data records,
identifying a record correlation inference of the plurality of record correlation inferences that has a confidence score greater than a pre-defined maximum confidence threshold, wherein identifying the record correlation inference that has the confidence score greater than the pre-defined maximum confidence threshold includes:
identifying a first respective record correlation inference of the plurality of record correlation inferences, and
using the first respective record correlation inference as the record correlation inference if a confidence score of the first respective record correlation inference is greater than confidence scores associated with a remainder of the plurality of record correlation inferences,
querying a computer database for a validated data record of the plurality of validated data records that is associated with the record correlation inference, and
setting the validated data record associated with the record correlation inference as the resolution candidate record for the unmapped data record; and
automatically re-assigning one or more data records of the computer database that are digitally associated with the unmapped data record to the resolution candidate record.
|