# US 2021/0342798 A1 · Performing enhanced deposit item processing using cognitive automation tools

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0342798 A1
- Document ID: US20210342798A1
- Status: Published application
- Kind: A1
- Application number: 16/865,759
- Filed: 2020-05-04
- Published: 2021-11-04
- Earliest claimed: 2020-05-04
- Assignees: Bank of America Corporation
- Applicants: Bank of America Corporation
- Inventors: Barry Hall; Adam Desautels; Verlon S. Watson, III; Letishia R. Hunt; Brendan P. Murphy; Charles Christopher Harbinson
- Web page: https://patentgrep.com/patent/US20210342798A1

## CPC

- G06Q 40/02: Banking, e.g. interest calculation or account maintenance (credit or loans G06Q40/03)
- G06F 16/55: Clustering; Classification
- G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F 18/24: Classification techniques
- G06F 40/205: Parsing
- G06N 5/04: Inference or reasoning models
- G06N 20/00: Machine learning
- G06Q 20/042: {characterized in that the payment protocol involves at least one cheque}
- G06Q 20/1085: {involving automatic teller machines [ATMs]}
- G06Q 30/0185: {Product, service or business identity fraud}
- G06V 30/19173: {Classification techniques}
- G06V 30/2253: {Recognition of characters printed with magnetic ink (G06V30/2247 takes precedence)}
- G06V 30/226: of cursive writing
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Aspects of the disclosure relate to performing enhanced deposit item processing using cognitive automation tools. In some embodiments, a computing platform may receive, from a deposit item input support server, image data associated with a deposit item. Subsequently, the computing platform may apply a machine learning classification model to the image data associated with the deposit item. In doing so, the computing platform may produce one or more predicted values for one or more fields of the deposit item and one or more confidence scores for the one or more predicted values. Based on these predicted values and confidence scores, the computing platform may generate and send one or more commands directing the deposit item input support server to accept or reject the deposit item, which in turn may cause the deposit item input support server to accept or reject the deposit item.

## Claims

### Claim 1

A computing platform, comprising:
at least one processor;
a communication interface communicatively coupled to the at least one processor; and
memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
receive, via the communication interface, from a first deposit item input support server, first image data associated with a first deposit item;
apply a machine learning classification model to the first image data associated with the first deposit item, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces one or more predicted values for one or more fields of the first deposit item and one or more confidence scores for the one or more predicted values;
based on the one or more predicted values for the one or more fields of the first deposit item and the one or more confidence scores for the one or more predicted values, generate one or more commands directing the first deposit item input support server to accept or reject the first deposit item; and
send, via the communication interface, to the first deposit item input support server, the one or more commands directing the first deposit item input support server to accept or reject the first deposit item, wherein sending the one or more commands directing the first deposit item input support server to accept or reject the first deposit item causes the first deposit item input support server to accept or reject the first deposit item.

### Claim 2 (depends on 1)

The computing platform of claim 1, wherein receiving the first image data associated with the first deposit item from the first deposit item input support server comprises receiving the first image data associated with the first deposit item from the first deposit item input support server in real time as the first deposit item is presented for deposit.

### Claim 3 (depends on 1)

The computing platform of claim 1, wherein receiving the first image data associated with the first deposit item from the first deposit item input support server comprises receiving image data comprising handwritten text.

### Claim 4 (depends on 3)

The computing platform of claim 3, wherein receiving the first image data associated with the first deposit item from the first deposit item input support server comprises parsing the handwritten text using one or more computer vision models.

### Claim 5 (depends on 1)

The computing platform of claim 1, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces a predicted value for a legal amount of the first deposit item and a confidence score for the predicted value for the legal amount of the first deposit item.

### Claim 6 (depends on 1)

The computing platform of claim 1, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces a predicted value for a convenience box amount of the first deposit item and a confidence score for the predicted value for the convenience box amount of the first deposit item.

### Claim 7 (depends on 1)

The computing platform of claim 1, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces a predicted value for a payee field of the first deposit item.

### Claim 8 (depends on 1)

The computing platform of claim 1, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces a predicted value for a date field of the first deposit item.

### Claim 9 (depends on 1)

The computing platform of claim 1, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces a predicted value for a signature field of the first deposit item.

### Claim 10 (depends on 1)

The computing platform of claim 1, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces a predicted value for an endorsement field of the first deposit item.

### Claim 11 (depends on 1)

The computing platform of claim 1, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces a predicted value for a magnetic ink character recognition (MICR) line field of the first deposit item.

### Claim 12 (depends on 1)

The computing platform of claim 1, wherein generating the one or more commands directing the first deposit item input support server to accept or reject the first deposit item comprises:
generating a valid-invalid decision for the first deposit item based on the one or more confidence scores for the one or more predicted values and one or more confidence thresholds corresponding to the one or more predicted values; and
embedding the valid-invalid decision generated for the first deposit item in the one or more commands directing the first deposit item input support server to accept or reject the first deposit item,
wherein the valid-invalid decision generated for the first deposit item causes the first deposit item input support server to accept the first deposit item if the valid-invalid decision generated for the first deposit item comprises an indication that the first deposit item is validly presented for deposit, and
wherein the valid-invalid decision generated for the first deposit item causes the first deposit item input support server to reject the first deposit item if the valid-invalid decision generated for the first deposit item comprises an indication that the first deposit item is not validly presented for deposit.

### Claim 13 (depends on 1)

The computing platform of claim 1, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
prior to receiving the first image data associated with the first deposit item, train the machine learning classification model based on a labeled dataset comprising images of previously submitted deposit items and decision data corresponding to the images of the previously submitted deposit items indicating whether each of the previously submitted deposit items was validly presented for deposit or not validly presented for deposit.

### Claim 14 (depends on 1)

The computing platform of claim 1, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
after sending the one or more commands directing the first deposit item input support server to accept or reject the first deposit item, receive, via the communication interface, from an enterprise user computing device, key input indicating one or more actual values for the one or more fields of the first deposit item; and
update the machine learning classification model based on the key input indicating the one or more actual values for the one or more fields of the first deposit item.

### Claim 15 (depends on 1)

The computing platform of claim 1, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
receive, via the communication interface, from a second deposit item input support server, second image data associated with a second deposit item, wherein the second deposit item input support server processes deposit items presented for deposit via a different channel than the first deposit item input support server;
apply the machine learning classification model to the second image data associated with the second deposit item, wherein applying the machine learning classification model to the second image data associated with the second deposit item produces one or more predicted values for one or more fields of the second deposit item and one or more confidence scores for the one or more predicted values of the one or more fields of the second deposit item;
based on the one or more predicted values for the one or more fields of the second deposit item and the one or more confidence scores for the one or more predicted values of the one or more fields of the second deposit item, generate one or more commands directing the second deposit item input support server to accept or reject the second deposit item; and
send, via the communication interface, to the second deposit item input support server, the one or more commands directing the second deposit item input support server to accept or reject the second deposit item, wherein sending the one or more commands directing the second deposit item input support server to accept or reject the second deposit item causes the second deposit item input support server to accept or reject the second deposit item.

### Claim 16

A method, comprising:
at a computing platform comprising at least one processor, a communication interface, and memory:
receiving, by the at least one processor, via the communication interface, from a first deposit item input support server, first image data associated with a first deposit item;
applying, by the at least one processor, a machine learning classification model to the first image data associated with the first deposit item, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces one or more predicted values for one or more fields of the first deposit item and one or more confidence scores for the one or more predicted values;
based on the one or more predicted values for the one or more fields of the first deposit item and the one or more confidence scores for the one or more predicted values, generating, by the at least one processor, one or more commands directing the first deposit item input support server to accept or reject the first deposit item; and
sending, by the at least one processor, via the communication interface, to the first deposit item input support server, the one or more commands directing the first deposit item input support server to accept or reject the first deposit item, wherein sending the one or more commands directing the first deposit item input support server to accept or reject the first deposit item causes the first deposit item input support server to accept or reject the first deposit item.

### Claim 17 (depends on 16)

The method of claim 16, wherein receiving the first image data associated with the first deposit item from the first deposit item input support server comprises receiving the first image data associated with the first deposit item from the first deposit item input support server in real time as the first deposit item is presented for deposit.

### Claim 18 (depends on 16)

The method of claim 16, wherein receiving the first image data associated with the first deposit item from the first deposit item input support server comprises receiving image data comprising handwritten text.

### Claim 19 (depends on 18)

The method of claim 18, wherein receiving the first image data associated with the first deposit item from the first deposit item input support server comprises parsing the handwritten text using one or more computer vision models.

### Claim 20

One or more non-transitory computer-readable media storing instructions that, when executed by a computing platform comprising at least one processor, a communication interface, and memory, cause the computing platform to:
receive, via the communication interface, from a first deposit item input support server, first image data associated with a first deposit item;
apply a machine learning classification model to the first image data associated with the first deposit item, wherein applying the machine learning classification model to the first image data associated with the first deposit item produces one or more predicted values for one or more fields of the first deposit item and one or more confidence scores for the one or more predicted values;
based on the one or more predicted values for the one or more fields of the first deposit item and the one or more confidence scores for the one or more predicted values, generate one or more commands directing the first deposit item input support server to accept or reject the first deposit item; and
send, via the communication interface, to the first deposit item input support server, the one or more commands directing the first deposit item input support server to accept or reject the first deposit item, wherein sending the one or more commands directing the first deposit item input support server to accept or reject the first deposit item causes the first deposit item input support server to accept or reject the first deposit item.
