US-12619490-B2 - System and method for electronic interaction recovery via multimodal machine learning models

Abstract

Systems, computer program products, and methods are described herein for electronic interaction recovery via multimodal machine learning models. The present disclosure is configured to receive an attempted electronic interaction from a set of communication channels; analyze the attempted electronic interaction from the set of communication channels associated with the attempted electronic interaction via a multimodal machine learning model (MLM) to detect a set of potential defects within the attempted electronic interaction, where the set of potential defects comprises indicators associated with encountered errors within the attempted electronic interaction; and trigger a recovery process of the attempted electronic interaction via the multimodal MLM.

Inventors

• Dinesh Kumar

Assignees

• BANK OF AMERICA CORPORATION

Dates

Publication Date

20260505

Application Date

20240326

Claims (18)

1. 1 . A system for electronic interaction recovery via multimodal machine learning models, the system comprising: at least one non-transitory storage device; and at least one processing device coupled to the at least one non-transitory storage device, wherein the at least one processing device is configured to: receive an attempted electronic interaction from a set of communication channels, wherein the set of communication channels comprises a set of application logs and a set of electronic communication records and wherein the set of electronic communication records comprises a set of image-based data; analyze the attempted electronic interaction from the set of communication channels associated with the attempted electronic interaction via a multimodal machine learning model (MLM) to detect a set of potential defects within the attempted electronic interaction, wherein analyzing the attempted electronic interaction from the set of communication channels associated with the attempted electronic interaction via a multimodal machine learning model (MLM) further comprises converting the set of electronic communication records into text-based data and wherein the set of potential defects comprises indicators associated with encountered errors within the attempted electronic interaction; and trigger a recovery process of the attempted electronic interaction upon identification of the set of potential defects within the attempted electronic interaction.
2. 2 . The system of claim 1 , wherein the set of electronic communication records further comprises a set of audio data.
3. 3 . The system of claim 1 , wherein the recovery process of the attempted electronic interaction via the multimodal MLM is an automated recovery.
4. 4 . The system of claim 1 , wherein the recovery process of the attempted electronic interaction via the multimodal MLM is a guided recovery process.
5. 5 . The system of claim 1 , wherein the set of communication channels further comprises interactive voice response calls.
6. 6 . The system of claim 1 , wherein the set of communication channels further comprises chatbot logs.
7. 7 . A computer program product for electronic interaction recovery via multimodal machine learning models, the computer program product comprising at least one non-transitory computer-readable medium having computer-readable program code portions embodied therein, the computer-readable program code portions which when executed by a processing device are configured to cause a processor to perform the following operations: receive an attempted electronic interaction from a set of communication channels, wherein the set of communication channels comprises a set of application logs and a set of electronic communication records and wherein the set of electronic communication records comprises a set of image-based data; analyze the attempted electronic interaction from the set of communication channels associated with the attempted electronic interaction via a multimodal machine learning model (MLM) to detect a set of potential defects within the attempted electronic interaction, wherein analyzing the attempted electronic interaction from the set of communication channels associated with the attempted electronic interaction via a multimodal machine learning model (MLM) further comprises converting the set of electronic communication records into text-based data and wherein the set of potential defects comprises indicators associated with encountered errors within the attempted electronic interaction; and trigger a recovery process of the attempted electronic interaction upon identification of the set of potential defects within the attempted electronic interaction.
8. 8 . The computer program product of claim 7 , wherein the set of electronic communication records further comprises a set of audio data.
9. 9 . The computer program product of claim 7 , wherein the recovery process of the attempted electronic interaction via the multimodal MLM is an automated recovery.
10. 10 . The computer program product of claim 7 , wherein the recovery process of the attempted electronic interaction via the multimodal MLM is a guided recovery process.
11. 11 . The computer program product of claim 7 , wherein the set of communication channels further comprises interactive voice response calls.
12. 12 . The computer program product of claim 7 , wherein the set of communication channels further comprises chatbot logs.
13. 13 . A computer-implemented method for electronic interaction recovery via multimodal machine learning models, the computer-implemented method comprising: receiving an attempted electronic interaction from a set of communication channels, wherein the set of communication channels comprises a set of application logs and a set of electronic communication records and wherein the set of electronic communication records comprises a set of image-based data; analyzing the attempted electronic interaction from the set of communication channels associated with the attempted electronic interaction via a multimodal machine learning model (MLM) to detect a set of potential defects within the attempted electronic interaction, wherein analyzing the attempted electronic interaction from the set of communication channels associated with the attempted electronic interaction via a multimodal machine learning model (MLM) further comprises converting the set of electronic communication records into text-based data and wherein the set of potential defects comprises indicators associated with encountered errors within the attempted electronic interaction; and triggering a recovery process of the attempted electronic interaction upon identification of the set of potential defects within the attempted electronic interaction.
14. 14 . The computer-implemented method of claim 13 , wherein the set of electronic communication records further comprises a set of audio data.
15. 15 . The computer-implemented method of claim 13 , wherein the recovery process of the attempted electronic interaction via the multimodal MLM is an automated recovery.
16. 16 . The computer-implemented method of claim 13 , wherein the recovery process of the attempted electronic interaction via the multimodal MLM is a guided recovery process.
17. 17 . The computer-implemented method of claim 13 , wherein the set of communication channels further comprises interactive voice response calls.
18. 18 . The computer-implemented method of claim 13 , wherein the set of communication channels further comprises chatbot logs.

Description

TECHNOLOGICAL FIELD Example embodiments of the present disclosure relate to electronic interaction recovery via multimodal machine learning models. BACKGROUND Incomplete or defective electronic interactions within an enterprise may cause delays and expenditure of resources to resolve. Analysis and recovery of these electronic interactions may be complicated further due to the plurality of mediums in which they may be generated. Applicant has identified a number of deficiencies and problems associated with electronic interaction recovery via multimodal machine learning models. Through applied effort, ingenuity, and innovation, many of these identified problems have been solved by developing solutions that are included in embodiments of the present disclosure, many examples of which are described in detail herein. BRIEF SUMMARY Systems, methods, and computer program products are provided for electronic interaction recovery via multimodal machine learning models. In one aspect, a system for electronic interaction recovery via multimodal machine learning models is presented. The system comprising a processing device, at least one non-transitory storage device, and at least one processing device coupled to the at least one non-transitory storage device wherein the at least one processing device is configured to: receive an attempted electronic interaction from a set of communication channels; analyze the attempted electronic interaction from the set of communication channels associated with the interaction via a multimodal machine learning model (MLM) to detect a set of potential defects within the attempted electronic interaction, wherein the set of potential defects comprises indicators associated with encountered errors within the attempted electronic interaction; and trigger a recovery process of the attempted electronic interaction upon identification of the set of potential defects within the attempted electronic interaction. In some embodiments, the set of communication channels comprise a set of application logs and a set of electronic communication records. In some embodiments, analyzing the set of communication channels associated with the attempted electronic interaction via the multimodal MLM further comprises converting the set of electronic communication records into text-based data. In some embodiments, the set of electronic communication records comprises a set of audio data. In some embodiments, the set of electronic communication records comprises a set of image-based data. In some embodiments, the recovery process of the attempted electronic interaction via the multimodal MLM is an automated recovery. In some embodiments, the recovery process of the attempted electronic interaction via the multimodal MLM is a guided recovery process. In another aspect, a computer program product for electronic interaction recovery via multimodal machine learning models is presented. The computer program product comprising at least one non-transitory computer-readable medium having computer-readable program code portions embodied therein, the computer-readable program code portions which when executed by a processing device are configured to cause the processor to perform the following operations: receive an attempted electronic interaction from a set of communication channels; analyze the attempted electronic interaction from the set of communication channels associated with the interaction via a multimodal machine learning model (MLM) to detect a set of potential defects within the attempted electronic interaction, wherein the set of potential defects comprises indicators associated with encountered errors within the attempted electronic interaction; and trigger a recovery process of the attempted electronic interaction upon identification of the set of potential defects within the attempted electronic interaction. In some embodiments, the set of communication channels comprise a set of application logs and a set of electronic communication records. In some embodiments, analyzing the set of communication channels associated with the attempted electronic interaction via the multimodal MLM further comprises converting the set of electronic communication records into text-based data. In some embodiments, the set of electronic communication records comprises a set of audio data. In some embodiments, the set of electronic communication records comprises a set of image-based data. In some embodiments, the recovery process of the attempted electronic interaction via the multimodal MLM is an automated recovery. In some embodiments, the recovery process of the attempted electronic interaction via the multimodal MLM is a guided recovery process. In another aspect, a computer-implemented method for electronic interaction recovery via multimodal machine learning models is presented. The computer-implemented method includes: receiving an attempted electronic interaction from a set of communication channels; analyzing the