# US 2021/0342124 A1 · Context-based recommendations for robotic process automation design

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

- Number: US 2021/0342124 A1
- Document ID: US20210342124A1
- Status: Published application
- Kind: A1
- Application number: 17/250,912
- Filed: 2019-09-26
- Published: 2021-11-04
- Earliest claimed: 2018-09-28
- Assignees: Element Ai Inc.
- Applicants: ELEMENT AI Inc.
- Inventors: Marie-Claude COTE; Alexei NORDELL-MARKOVITS; Andrej TODOSIC
- Web page: https://patentgrep.com/patent/US20210342124A1

## CPC

- G06F 8/20: Software design
- G06F 8/33: Intelligent editors
- G06N 3/09: Supervised learning
- G06N 5/04: Inference or reasoning models
- G06N 20/00: Machine learning
- G06N 3/045: Combinations of networks
- G06N 3/08: Learning methods
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Systems and methods for adding process actions to the design of a robotic software process. A context-recognition module recognizes a current state of a process being designed, and passes information on that current state to a recommendation module. The recommendation module evaluates the current state and identifies at least one suitable process action to recommend in response to that current state. The recommendation module then recommends the at least one process action to the human designer. If the designer accepts the recommendation, a design module adds the process action to the process design. The recommendation module may also use information about previous actions in the process and in other processes when identifying suitable process actions. The context-recognition module and the recommendation module may each comprise at least one machine learning module, which may or may not be neural network based.

## Claims

### Claim 1

A method for adding at least one process action to a design of a software process, the method comprising:
(a) recognizing a current state in said software process;
(b) based on said current state, making a recommendation of said at least one process action to a designer of said software process;
(c) receiving a response to said recommendation from said designer; and
(d) adding said at least one process action to said design when said response is an acceptance of said at least one process action,
wherein machine learning is used in making said recommendation in step (b).

### Claim 2 (depends on 1)

The method according to claim 1, wherein said current state is related to an external software application.

### Claim 3 (depends on 1)

The method according to claim 1, further comprising a plurality of process actions, wherein all of said plurality of process actions are recommended to said user simultaneously.

### Claim 4 (depends on 1)

The method according to claim 1, further comprising a plurality of process actions, wherein each of said plurality of process actions is recommended to said user sequentially.

### Claim 5 (depends on 1)

The method according to claim 1, wherein said at least one process action comprises multiple sub-actions.

### Claim 6 (depends on 1)

The method according to claim 1, wherein said current state is related to a current action in said software process.

### Claim 7 (depends on 1)

The method according to claim 1, wherein said recommending in step (b) is further based on at least one previous action in said software process.

### Claim 8

(canceled)

### Claim 9 (depends on 1)

The method according to claim 1, wherein said at least one process action is at least one of:
a data manipulation action;
an action related to an interaction between systems;
a communication-related action; and
a storage-related action.

### Claim 10 (depends on 1)

The method of claim 1, further comprising the steps of:
(e) collecting data related to: said current state; said at least one process action; and said response; and
(f) using machine learning with said data to refine future recommendations.

### Claim 11

A system for adding at least one process action to a design of a software process, the system comprising:
a context-recognition module for recognizing a current state of said software process;
a machine learning enabled recommendation module for:
making an evaluation of said current state;
based on said evaluation, determining said at least one process action to recommend;
making a recommendation of said at least one process action to a designer of said software process; and
receiving a response to said recommendation from said designer; and
a design module for adding said at least one process action to said design when said response is an acceptance of said process action.

### Claim 12 (depends on 11)

The system according to claim 11, wherein said current state is related to an external software application.

### Claim 13 (depends on 11)

The system according to claim 11, further comprising a plurality of process actions, wherein all of said plurality of process actions are recommended to said user simultaneously.

### Claim 14 (depends on 11)

The system according to claim 11, further comprising a plurality of process actions, wherein each of said plurality of process actions are recommended to said user sequentially.

### Claim 15 (depends on 11)

The system according to claim 11, wherein said at least one process action comprises multiple process actions.

### Claim 16 (depends on 11)

The system according to claim 11, wherein said current state is related to a current action in said software process.

### Claim 17 (depends on 11)

The system according to claim 11, further comprising a history module for storing information related to a previous action in said software process, and wherein said recommendation module uses said information in determining said at least one process action to recommend.

### Claim 18 (depends on 17)

The system according to claim 17, wherein said history module further stores other information related to other software processes, and wherein said other information is also used by said recommendation module in determining said at least one process action to recommend.

### Claim 19 (depends on 11)

The system according to claim 11, wherein said at least one process action is at least one of:
a data manipulation action;
an action related to an interaction between systems;
a communication-related action; and
a storage-related action.

### Claim 20 (depends on 17)

The system of claim 17, wherein said history module further stores data related to: said current state; said at least one process action; and said response, and wherein said recommendation module uses said data to refine future recommendations.

### Claim 21

Non-transitory computer-readable media having encoded thereon computer-readable and computer-executable instructions that, when executed, implement a method for adding at least one process action to a design of a software process, the method comprising:
(a) recognizing a current state in said software process;
(b) based on said current state, making a recommendation of said at least one process action to a designer of said software process;
(c) receiving a response to said recommendation from said designer; and
(d) adding said at least one process action to said design when said response is an acceptance of said at least one process action,
wherein machine learning techniques are used in making said recommendation in step (b).
