IP Library Granted Patent US 11,361,567
Granted Patent B2
US 11,361,567 · App. 17/142,758 · Granted Jun 14, 2022

Methods and systems for facilitating classification of portions of a regulatory document

Inventors: Kayvan Alikhani (Berkeley, CA); Chuan Yun (Santa Clara, CA); Mohit Mohan Galvankar (Alameda, CA); Mohsen Emaminouri (San Francisco, CA)
Assignee: Catachi Co.
G06V30/413G06F16/285G06F16/93G06F40/103G06F40/169G06K9/6256G06N20/00
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,361,567
App. No.
17/142,758
Granted
Jun 14, 2022
Kind
B2
Abstract

Disclosed herein is a method for facilitating classification of portions of a regulatory document. Accordingly, the method may include receiving, using a communication device, the regulatory document from a device, converting, using a processing device, the regulatory document from a source format to a preferred format, analyzing, using the processing device, the regulatory document of the preferred format based on the converting, identifying, using the processing device, the portions of the regulatory document based on the analyzing, and classifying, using the processing device, the portions using a classifier model into classification codes based on the identifying. Further, the method may include generating, using the processing device, an annotated regulatory document based on the classifying. Further, the method may include transmitting, using the communication device, the annotated regulatory document to the device. Further, the method may include storing, using a storage device, the annotated regulatory document and the classifier model.

Claims (71)

1. A method for facilitating classification of portions of a regulatory document, the method comprising:

receiving, using a communication device, the regulatory document from at least one device, wherein the regulatory document is associated with a structure, wherein the structure comprises a layout of content of the regulatory document, wherein the layout comprises at least one of an indent and a white space associated with the portions of the regulatory document, wherein the regulatory document is associated with a source format;

converting, using a processing device, the regulatory document from the source format to a preferred format;

analyzing, using the processing device, the regulatory document of the preferred format based on the converting;

identifying, using the processing device, a plurality of portions of the regulatory document based on the analyzing;

classifying, using the processing device, the plurality of portions using a classifier model into a plurality of classification codes based on the identifying, wherein the classifying of the plurality of portions is based on the at least one of the indent and the white space associated with the plurality of portions;

generating, using the processing device, a visually annotated regulatory document based on the classifying, wherein the annotated regulatory document comprises the plurality of portions and the plurality of classification codes associated with the plurality of portions;

transmitting, using the communication device, the annotated regulatory document to the at least one device;

storing, using a storage device, the annotated regulatory document and the classifier model;

analyzing, using the processing device, the annotated regulatory document based on the generating;

identifying, using the processing device, the plurality of classification codes associated with the plurality of portions based on the analyzing of the annotated regulatory document;

generating, using the processing device, a plurality of visual indicators for the plurality of classification codes based on the identifying of the plurality of classification codes;

applying, using the processing device, the plurality of visual indicators to the plurality of portions based on the plurality of classification codes of the plurality of portions;

generating, using the processing device, a visually annotated regulatory document based on the applying;

transmitting, using the communication device, the visually annotated regulatory document to the at least one device, wherein the at least one device is configured for presenting the visually annotated regulatory document;

receiving, using the communication device, at least one feedback associated with at least one classification code of at least one portion of the plurality of portions from the at least one device based on the transmitting of the visually annotated regulatory document;

modifying, using the processing device, the at least one classification code of the at least one portion based on the at least one feedback;

generating, using the processing device, at least one modified classification code for the at least one portion based on the modifying;

generating, using the processing device, a modified annotated regulatory document based on the generating of the at least one modified classification code;

storing, using the storage device, the modified annotated regulatory document;

receiving, using the communication device, at least one first feedback associated with the annotated regulatory document from the at least one device based on the transmitting of the annotated regulatory document;

analyzing, using the processing device, the at least one first feedback;

modifying, using the processing device, the classifier model based on the analyzing of the at least one first feedback;

generating, using the processing device, a refined classifier model based on the modifying, wherein the refined classifier model is configured for classifying the regulatory document with a higher classification accuracy as compared to a classifying accuracy of the classifier model; and

storing, using the storage device, the refined classifier model.

2. The method of claim 1 , wherein the plurality of classification codes comprises at least two of a title, a section heading, a section, a sub-section heading, and a sub-section.

3. The method of claim 1 further comprising training, using the processing device, the classifier model based on a sample corpus of regulatory documents, wherein the classifier model comprises a machine learning model, wherein the classifier model is configured for classifying the plurality of portions based on the training.

4. The method of claim 1 further comprising retraining, using the processing device, the classifier model based on the at least one feedback, wherein the classifier model is configured for classifying the plurality of portions based on the retraining.

5. The method of claim 1 further comprising:

receiving, using the communication device, a plurality of regulatory document corpus from a plurality of document source devices;

analyzing, using the processing device, the plurality of regulatory document corpus;

generating, using the processing device, a plurality of classifier models based on the analyzing of the plurality of regulatory document corpus; and

storing, using the storage device, the plurality of classifier models.

6. The method of claim 5 further comprising identifying, using the processing device, the classifier model of the plurality of classifier models based on the analyzing of the regulatory document, wherein the classifying of the plurality of portions using the classifier model into the plurality of classification codes is further based on the identifying of the classifier model.

7. The method of claim 1 , wherein the white space comprises one or more characters representing at least one of a horizontal space and a vertical space in a typography of the regulatory document.

8. A system for facilitating classification of portions of a regulatory document, the system comprising:

a communication device;

a processing device;

a storage device;

the system being configured for:

receiving, using the communication device, the regulatory document from at least one device, wherein the regulatory document is associated with a structure, wherein the structure comprises a layout of content of the regulatory document, wherein the layout comprises at least one of an indent and a white space associated with the portions of the regulatory document, wherein the regulatory document is associated with a source format;

converting, using the processing device, the regulatory document from the source format to a preferred format;

analyzing, using the processing device, the regulatory document of the preferred format based on the converting;

identifying, using the processing device, a plurality of portions of the regulatory document based on the analyzing;

classifying, using the processing device, the plurality of portions using a classifier model into a plurality of classification codes based on the identifying, wherein the classifying of the plurality of portions is based on the at least one of the indent and the white space associated with the plurality of portions;

generating, using the processing device, a visually annotated regulatory document based on the classifying, wherein the annotated regulatory document comprises the plurality of portions and the plurality of classification codes associated with the plurality of portions;

transmitting, using the communication device, the annotated regulatory document to the at least one device;

storing, using the storage device, the annotated regulatory document and the classifier model;

analyzing, using the processing device, the annotated regulatory document based on the generating;

identifying, using the processing device, the plurality of classification codes associated with the plurality of portions based on the analyzing of the annotated regulatory document;

generating, using the processing device, a plurality of visual indicators for the plurality of classification codes based on the identifying of the plurality of classification codes;

applying, using the processing device, the plurality of visual indicators to the plurality of portions based on the plurality of classification codes of the plurality of portions;

generating, using the processing device, a visually annotated regulatory document based on the applying;

transmitting, using the communication device, the visually annotated regulatory document to the at least one device, wherein the at least one device is configured for presenting the visually annotated regulatory document;

receiving, using the communication device, at least one feedback associated with at least one classification code of at least one portion of the plurality of portions from the at least one device based on the transmitting of the visually annotated regulatory document;

modifying, using the processing device, the at least one classification code of the at least one portion based on the at least one feedback;

generating, using the processing device, at least one modified classification code for the at least one portion based on the modifying;

generating, using the processing device, a modified annotated regulatory document based on the generating of the at least one modified classification code;

storing, using the storage device, the modified annotated regulatory document;

wherein the communication device is further configured for receiving at least one first feedback associated with the annotated regulatory document from the at least one device based on the transmitting of the annotated regulatory document, wherein the processing device is further configured for:

analyzing the at least one first feedback;

modifying the classifier model based on the analyzing of the at least one first feedback; and

generating a refined classifier model based on the modifying, wherein the refined classifier model is configured for classifying the regulatory document with a higher classification accuracy as compared to a classifying accuracy of the classifier model, wherein the storage device is further configured for storing the refined classifier model.

9. The system of claim 8 , wherein the plurality of classification codes comprises at least two of a title, a section heading, a section, a sub-section heading, and a sub-section.

10. The system of claim 8 , wherein the processing device is further configured for training the classifier model based on a sample corpus of regulatory documents, wherein the classifier model comprises a machine learning model, wherein the classifier model is configured for classifying the plurality of portions based on the training.

11. The system of claim 8 , wherein the processing device is further configured for retraining the classifier model based on the at least one feedback, wherein the classifier model is configured for classifying the plurality of portions based on the retraining.

12. The system of claim 8 , wherein the communication device is further configured for receiving a plurality of regulatory document corpus from a plurality of document source devices, wherein the processing device is further configured for:

analyzing the plurality of regulatory document corpus; and

generating a plurality of classifier models based on the analyzing of the plurality of regulatory document corpus, wherein the storage device is further configured for storing the plurality of classifier models.

13. The system of claim 12 , wherein the processing device is further configured for identifying the classifier model of the plurality of classifier models based on the analyzing of the regulatory document, wherein the classifying of the plurality of portions using the classifier model into the plurality of classification codes is further based on the identifying of the classifier model.

14. The system of claim 8 , wherein the white space comprises one or more characters representing at least one of a horizontal space and a vertical space in a typography of the regulatory document.

Assignments (3)
MERGER Recorded Oct 18, 2024
From: CATACHI CO. DBA COMPLIANCE.AI
To: ARCHER TECHNOLOGIES LLC
Reel/Frame 068945/0870 →
PATENT SHORT FORM SECURITY AGREEMENT Recorded Apr 8, 2024
From: CATACHI CO.
To: GOLUB CAPITAL MARKETS LLC, AS COLLATERAL AGENT
Reel/Frame 067040/0258 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 6, 2021
From: ALIKHANI, KAYVAN; YUN, CHUAN; GALVANKAR, MOHIT MOHAN; EMAMINOURI, MOHSEN
To: CATACHI CO. DBA COMPLIANCE.AI
Reel/Frame 054830/0487 →
Continuity (2)
Provisional Application 62957595 · Jan 6, 2020
Related Publication 20210209355A1 · Jul 8, 2021