Specification mapping and message correction for reducing message rejections

A method for message specification mapping to reduce message rejections, the method comprising:

• training a machine learning component of a message correction manager using successful transaction requests;
• receiving a transaction request generated by a requester device, the transaction request comprising a message format, a plurality of data fields, and message data;
• analyzing the transaction request using a message specification associated with a recipient device of the transaction request prior to a first submission attempt to the recipient device from the requester device, the message specification describing a set of rules associated with the message format, the plurality of data fields, and the message data required for acceptance of a message by the recipient device;
• automatically identifying, using the trained machine learning component of the message correction manager, a noncompliant message element associated with the transaction request based on the analysis of the transaction request using the message specification the analysis comprising applying the set of rules to the message format, the plurality of data fields, and the message data available within the transaction request, the noncompliant message element comprising a portion of the transaction request that fails to comply with the message specification;
• automatically correcting, using the trained machine learning component of the message correction manager, the noncompliant message element to form a corrected transaction request;
• and
• submitting the corrected transaction request to the recipient device at the first submission attempt via a network, wherein the corrected transaction request complies with the message specification of the recipient device to prevent rejection of the transaction request by the recipient device based on noncompliance with the message specification;
• generating an audit log comprising:
• corrections to the transaction request made by the trained machine learning component of the message correction manager, a date and time the transaction request was analyzed, corrected, and submitted to the recipient device, and a listing of errors identified in the transaction request;
• and
• providing at least a summary of the audit log to a user enabling the user to adjust a transaction request generation process to reduce a number of transactions requests requiring correction.

A computer storage medium having computer-executable instructions stored thereon, which, upon execution by a computer, cause the computer to perform operations comprising:

• training a machine learning component of a message correction manager using successful transaction requests;
• mapping a transaction request generated by a user device to a message specification associated with a recipient device, the message specification describing a set of rules for acceptance of a message by the recipient device, the set of rules including one or more of:
• a message format, a message data limit, a plurality of data fields, and message data;
• automatically identifying, using the trained machine learning component of the message correction manager, a noncompliant message element associated with the transaction request based on the mapping of the transaction request to the message specification, the mapping including analyzing the transaction request by applying the set of rules to the message format, the plurality of data fields, and the message data available within the transaction request;
• automatically correcting, using the trained machine learning component of the message correction manager, the noncompliant message element to conform with the message specification of the recipient device without a rejection of the transaction request by the recipient device;
• submitting a corrected transaction request to the recipient device via a network without having first received a rejection of the transaction request by the user device from the recipient device based on noncompliance with the message specification;
• generating an audit log comprising:
• corrections to the transaction request made by the trained machine learning component of the message correction manager, a date and time the transaction request was analyzed, corrected, and submitted to the recipient device, and a listing of errors identified in the transaction request;
• and
• providing at least a summary of the audit log to a user enabling the user to adjust a transaction request generation process to reduce a number of transactions requests requiring correction.