Method and system for artificial intelligence based enhancement of sale processes

Patent metadata

Field	Value
Publication number	US-12579576-B2
Application number	US-202318535297-A
Country	US
Kind code	B2
Filing date	Dec 11, 2023
Priority date	Dec 11, 2023
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

The method includes: receiving a communication that relates to a request for a price quote for at least one equity derivative product; extracting from the communication, attributes of the requested price quote for the at least one equity derivative product; generating based on the extracted attributes, a template request that has a predetermined format for each of the at least one equity derivative product; displaying, via a graphical user interface (GUI), each template request for review by a user, wherein the GUI includes an input mechanism for at least one of accepting and modifying each template request; transmitting each reviewed template request to a pricing system; receiving, from the pricing system, a quote for each of the at least one equity derivative product; and displaying, by the GUI, the price quote.

First claim

The invention claimed is:

1 . A method for automatically formatting communications into a structured format, the method being implemented by at least one processor, the method comprising:
receiving, by the at least one processor via a front-end application, a communication that relates to a request for a price quote for at least one equity derivative product, wherein the front-end application contains logic for recording and logging a user action associated with the request for the price quote; transmitting, by the at least one processor, the communication from the front-end application to a natural-language-processing (NLP) machine-learning (ML) model; mediating, by the at least one processor via a central parsing service, the transmitting of the communication, wherein the mediating includes recording and logging information during the transmitting of the communication, wherein the central parsing service includes natural language understanding (NLU) components for template request generation, and wherein the central parsing service contains logic for transforming the communication into a shorthand template; extracting, by the at least one processor via the NLP-ML model from the shorthand template, attributes of the requested price quote for the at least one equity derivative product; performing, by the at least one processor via the NLP-ML model, at least one verification operation on the extracted attributes; transmitting, by the at least one processor, the extracted attributes to the central parsing service; generating, by the at least one processor via the central parsing service, a template request based on the extracted attributes, wherein the template request has a predetermined format for each of the at least one equity derivative product, and wherein the central parsing service automatically fills in for missing attributes during the generating of the template request; transmitting, by the at least one processor, the template request from the central parsing service to the front-end application; simplifying, by the at least one processor via the front-end application, the template request by applying parameters to the request for the price quote; displaying, by the at least one processor via a graphical user interface (GUI) of the front-end application, each simplified template request for review by a user, wherein the GUI includes an input mechanism for at least one from among accepting and modifying each template request; transmitting, by the at least one processor, information received by the input mechanism to a feedback application programming interface (API) for implementing improvements for automatically formatting communications into a structured format; transmitting, by the at least one processor, each reviewed template request to a pricing system; receiving, by the at least one processor via the front-end application from the pricing system, a price quote for each of the at least one equity derivative product; and displaying, by the GUI, the price quote.

2 . The method of claim 1 , wherein the extracting of the attributes comprises:
applying the NLP-ML model to identify each of an asset, an expiration date, a strike price, a type of equity derivative product, and a number of units from the communication; and formatting each of the asset, the expiration date, the strike price, the type of equity derivative product, and the number of units into the template request having the predetermined format.

3 . The method of claim 1 , wherein the attributes comprise at least one from among assets, dates, types of requested equity derivative products, strike prices, units, expiration dates, notionals, and currencies.

4 . The method of claim 1 , wherein the communication comprises at least one from among an un-structured text request, a colloquial terminology request, and a natural language request, and wherein the communication is received via at least one from among a voice command, an audio command, a text command, an email command, and an electronic communication command.

5 . The method of claim 1 , wherein the at least one equity derivative product includes textual information relating to at least one from among an option, a put, a call, a future, a warrant, a swap, a single-leg option, a put spread, a call spread, call vs call, a barrier option, and a multi-leg option.

6 . The method of claim 1 , wherein the modifying of each template request comprises prompting the user to provide an input for at least one from among adding at least one attribute parameter, removing at least one attribute parameter, and altering at least one attribute parameter from each template request.

7 . The method of claim 1 , wherein the extracting of the attributes comprises:
analyzing, by the at least one processor via the NLP-ML model, the communication to identify at least one sequence of characters and a respective attribute category that is included in a list of attribute categories that is associated with the predetermined format; and wherein the generating of the template request comprises designating each of the identified at least one sequence of characters in order by their respective attribute category to match the predetermined format.

8 . The method of claim 7 , wherein the generating of the template request further comprises applying a sequence of natural language understanding (NLU) routines for performing the designating of each of the identified at least one sequence of characters, and
when at least one attribute from the predetermined format is missing from the extracted attributes, the method further comprises generating a default attribute value to populate the template request to match the predetermined format.

9 . The method of claim 1 , wherein the central parsing service includes:
an NLP service request formatter, wherein the NLP service request formatter transforms the communication into a first format that is compatible with the NLP-ML model; a format command parser, wherein the format command parser transforms the extracted attributes into a command line representation shorthand; and a details logger, wherein the details logger performs the recording and logging of the information.

10 . The method of claim 1 , wherein the NLP-ML model includes:
an artificial intelligence (AI) ML named entity recognizer (NER) that processes the communication into a sequence of characters that are transmitted to a deep learning model to generate the template request based on a character embedding and a word-level feature embedding.

11 . A computing apparatus for automatically formatting communications into a structured format, the computing apparatus comprising:
a processor; a memory; a display; and a communication interface coupled to each of the processor, the memory, and the display, wherein the processor is configured to: receive, via the communication interface and a front-end application, a communication that relates to a request for a price quote for at least one equity derivative product, wherein the front-end application contains logic for recording and logging a user action associated with the request for the price quote; transmit the communication from the front-end application to a natural-language-processing (NLP) machine-learning (ML) model; mediate, via a central parsing service, the transmitting of the communication, wherein the mediating includes recording and logging information during the transmitting of the communication, wherein the central parsing service includes natural language understanding (NLU) components for template request generation, and wherein the central parsing service contains logic for transforming the communi

Assignees

• Jpmorgan Chase Bank Na

Inventors

• Marold Jake
• Dai Mike
• Vossos Harlan
• Cruikshanks Stuart
• Peak William
• Yoong Guo-Zheng

Classifications

• G06Q40/045

  Accepting or processing orders in an exchange · CPC title

• G06Q40/0431

  Displaying financial market data · CPC title

• G06Q40/0421

  using artificial intelligence, machine learning or neural networks · CPC title

• G06Q40/04

  Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange · CPC title

• G06N20/00 Primary

  Machine learning · CPC title