Granted Patent B2

US 12,651,273 · App. 18/491,175 · Granted Jun 9, 2026

Inventors: Ashish Agrawal (Hyderabad, IN); Miltiadis Mitrakas (Essex, GB); Ian Yi Seaw (Singapore, SG); Yu Tsuneoka (Kanagawa-Ken, JP); Alex Astle (Jersey City, NJ); Rahul Kalia (Hong Kong, HK); Riya Ojha (Hyderabad, IN)

Assignee: JPMORGAN CHASE BANK, N.A.

G06Q30/0203G06F40/103G06V30/191G06V30/30G06V30/41

Abstract

Claims (58)

1 . A method for questionnaire data digitization and reconciliation by utilizing one or more processors along with allocated memory, the method comprising:

implementing an automated email tool into a data flow pipeline, wherein the email tool generates an autonomous program for continuously monitoring a shared mailbox for unread emails originated from a computing system, wherein the computing system is utilized for reviewing a document including questionnaire data containing a plurality of line items filled out by a client utilizing a client computing device;

fetching, in response to determining that the shared mailbox contains unread emails, the document as attached to an unread email and storing the document in a shared drive;

accessing the document from the shared drive; implementing an optical character recognition (OCR) tool into the data flow pipeline;

converting, using the OCR tool, the questionnaire data containing the plurality of line items into machine-readable format data;

implementing an automated reconciliation tool into the data flow pipeline;

reading, by the automated reconciliation tool, the machine-readable format data for each line item;

comparing, by the automated reconciliation tool, data for each line item against a corresponding predefined guidance data;

identifying, based on comparing, by the automated reconciliation tool, missing response data, negative response data, and insufficient response data corresponding to the questionnaire data filled out by the client by applying predefined rules; and

automatically reconciling the missing response data, negative response data, and insufficient response data, wherein the automatically reconciling comprises:

filtering the missing response data from the questionnaire data and grouping the missing response data into a first file;

filtering the negative response data from the questionnaire data and grouping the negative response data into a second file; and

filtering the insufficient response data from the questionnaire data and grouping the insufficient response data into a third file;

transmitting the first file, the second file, and the third file as output to the client computing device via a communication interface;

receiving input data from the client corresponding to each of the missing response data, the negative response data, and the insufficient response data; and

repeating the converting, the reading, the comparing, the identifying, and the automatically reconciling on the received input data,

wherein the automated reconciliation tool and the OCR tool are implemented as part of a data digitization and reconciliation module that is platform, language, database, browser, and cloud agnostic, thereby enabling orchestration and passing of the questionnaire data through the data flow pipeline across various components, and wherein configuration or data files utilized by the data digitization and reconciliation module are stored in a configuration-based language format such that the data digitization and reconciliation module is tunable or modifiable for performance of without affecting the configuration or data files.

2 . The method according to claim 1 , wherein the questionnaire data includes Correspondent Banking Due Diligence Questionnaire (CBDDQ) data.

3 . The method according to claim 1 , further comprising: implementing an artificial intelligence and machine learning algorithm into the automated reconciliation tool for automatically reconciling the missing response data, negative response data, and insufficient response.

4 . A system for questionnaire data digitization and reconciliation, the system comprising:

a processor; and

a memory operatively connected to the processor via a communication interface, the memory storing computer readable instructions, when executed, causes the processor to:

implement an automated email tool into a data flow pipeline, wherein the email tool generates an autonomous program for continuously monitoring a shared mailbox for unread emails originated from a computing system, wherein the computing system is utilized for reviewing a document including questionnaire data containing a plurality of line items filled out by a client utilizing a client computing device;

fetch, in response to determining that the shared mailbox contains unread emails, the document as attached to an unread email and store the document in a shared drive;

access the document from the shared drive;

implement an optical character recognition (OCR) tool into the data flow pipeline;

convert, using the OCR tool, the questionnaire data containing the plurality of line items into machine-readable format data;

implement an automated reconciliation tool into the data flow pipeline;

read, using the automated reconciliation tool, the machine-readable format data for each line item;

compare, using the automated reconciliation tool, data for each line item against a corresponding predefined guidance data;

identify, based on the comparing, missing response data, negative response data, and insufficient response data corresponding to the questionnaire data by applying predefined rules; and

automatically reconcile the missing response data, negative response data, and insufficient response data, wherein automatically reconciling comprises:

filtering the missing response data from the questionnaire data and group the missing response data into a first file;

filtering the negative response data from the questionnaire data and group the negative response data into a second file; and

filtering the insufficient response data from the questionnaire data and group the insufficient response data into a third file;

transmit the first file, the second file, and the third file as output to the client computing device via a communication interface;

receive input data from the client corresponding to each of the missing response data, the negative response data, and the insufficient response data; and

repeat the converting, the reading, the comparing, the identifying, and the automatically reconciling on the received input data,

wherein the automated reconciliation tool and the OCR tool are implemented as part of a data digitization and reconciliation module that is platform, language, database, browser, and cloud agnostic, thereby enabling orchestration and passing of the questionnaire data through the data flow pipeline across various components, and wherein configuration or data files utilized by the data digitization and reconciliation module are stored in a configuration-based language format such that the data digitization and reconciliation module is tunable or modifiable for performance without affecting the configuration or data files.

5 . The system according to claim 4 , wherein the questionnaire data includes Correspondent Banking Due Diligence Questionnaire (CBDDQ) data.

6 . The system according to claim 4 , wherein the processor is further configured to: implement an artificial intelligence and machine learning algorithm into the automated reconciliation tool for automatically reconciling the missing response data, negative response data, and insufficient response.

7 . A non-transitory computer readable medium configured to store having instructions stored thereon, wherein the instructions are executable to cause a processor to perform operations comprising:

implementing an automated email tool into a data flow pipeline, wherein the email tool generates an autonomous program for continuously monitoring a shared mailbox for unread emails originated from a computing system, wherein the computing system is utilized for reviewing a document including questionnaire data containing a plurality of line items filled out by a client utilizing a client computing device;

fetching, in response to determining that the shared mailbox contains unread emails, the document as attached to an unread email and storing the document in a shared drive;

accessing the document from the shared drive;

implementing an optical character recognition (OCR) tool into the data flow pipeline;

converting the questionnaire data containing the plurality of line items into machine-readable format data; implementing an automated reconciliation tool into the data flow pipeline; reading, by the automated reconciliation tool, the machine-readable format data for each line item;

comparing data for each line item against a corresponding predefined guidance data;

identifying, based on the comparing, missing response data, negative response data, and insufficient response data corresponding to the questionnaire data by applying predefined rules; and

automatically reconciling the missing response data, negative response data, and insufficient response data, wherein the automatically reconciling comprises:

filtering the missing response data from the questionnaire data and grouping the missing response data into a first file:

filtering the negative response data from the questionnaire data and grouping the negative response data into a second file; and

filtering the insufficient response data from the questionnaire data and grouping the insufficient response data into a third file;

transmitting the first file, the second file, and the third file as output to the client computing device via a communication interface;

receiving input data from the client corresponding to each of the missing response data, the negative response data, and the insufficient response data; and

repeating the converting, the reading, the comparing, the identifying, and the automatically reconciling on the received input data,

wherein the automated reconciliation tool and the OCR tool are implemented as part of a data digitization and reconciliation module that is platform, language, database, browser, and cloud agnostic, thereby enabling orchestration and passing of the questionnaire data through the data flow pipeline across various components, and wherein configuration or data files utilized by the data digitization and reconciliation module are stored in a configuration-based language format such that the data digitization and reconciliation module is tunable or modifiable for performance without affecting the configuration or data files.

8 . The non-transitory computer readable medium according to claim 7 , wherein the questionnaire data includes Correspondent Banking Due Diligence Questionnaire (CBDDQ) data and, wherein the operations further comprise: implementing an artificial intelligence and machine learning algorithm into the automated reconciliation tool for automatically reconciling the missing response data, negative response data, and insufficient response.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (16)