# US 2022/0044579 A1 · Training system for training a user in a predefined operation

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

- Number: US 2022/0044579 A1
- Document ID: US20220044579A1
- Status: Published application
- Kind: A1
- Application number: 17/398,411
- Filed: 2021-08-10
- Published: 2022-02-10
- Earliest claimed: 2020-08-10
- Assignees: Play Games 24x7 Pvt. Ltd.
- Applicants: Play Games 24x7 Pvt. Ltd.
- Inventors: Sharanya Eswaran; Mridul Sachdeva; Tridib Mukherjee; Vikram Vimal; Deepanshi Seth; Sanjay Kumar Agrawal
- Web page: https://patentgrep.com/patent/US20220044579A1

## CPC

- G09B 5/02: with visual presentation of the material to be studied, e.g. using film strip
- G06Q 20/405: {Establishing or using transaction specific rules}
- G07F 17/3239: {Tracking of individual players}
- G06N 20/00: Machine learning
- G09B 19/22: Games, e.g. card games
- G09B: EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS (devices for psychotechnics or for testing reaction times A61B5/16; games, sports, amusements A63; projectors, projector screens G03B)
- 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
- G07F: COIN-FREED OR LIKE APPARATUS (coin sorting G07D3/00; coin testing G07D5/00; {handling coins or paper currencies apart from payment activated apparatus G07D; payment architectures, schemes or protocols G06Q20/00})
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A training system is disclosed here for training a user in a predefined operation to make a best decision in a predefined operation. A first storage module stores a first log of the decisions made by the skilled users during the predefined operation. A first pre-processing module is in communication with the first storage module to pre-process the first log to generate a first multi-dimensional image array of the predefined operation. A training module trains a model based on the first multi-dimensional image array to generate a skilled strategy model. A second storage module stores a second log of decisions made by the user in the predefined operation and pre-processes the second log to generate a second multi-dimensional image array. A comparison module compares the second multi-dimensional image array with the skilled strategy model to generate a prediction of the best decision to be made by the user.

## Claims

### Claim 1

A training method for training a user performing a predefined operation to make a best decision based on previous decisions made by skilled users in the predefined operation, the training method comprising:
processing stored instructions in a first storage module and a second storage module via at least one processor coupled with the first storage module and the second storage module, the stored instructions comprising:
storing; via the first storage module, a first log of the decisions made by the skilled users during the predefined operation;
pre-processing, via a first pre-processing module in communication with the first storage module, the first log to generate a first multi-dimensional image array of each state of the predefined operation;
training, via a training module, a model based on the first log at a predetermined state of the predefined operation that is captured as the first multi-dimensional image array to generate a skilled strategy model;
storing, via the second storage module, a second log of decisions made by the user in the predefined operation;
pre-processing, via a second pre-processing module in communication with the second storage module, the second log to generate a second multi-dimensional image array; and
comparing, via a comparison module, the second multi-dimensional image array with the skilled strategy model to generate a prediction of the best decision to be made by the user.

### Claim 2 (depends on 1)

The training method as claimed in claim 1, wherein the first pre-processing module generates snapshots based on the pre-processed first log to make the first log machine readable and machine learnable.

### Claim 3 (depends on 1)

The training method as claimed in claim 1, wherein the prediction of best move is displayed on a user device being used by the user so that the user is prompted to execute the best move.

### Claim 4 (depends on 1)

The training method as claimed in claim 1, wherein the training module comprises a personalized upskilling module that performs personalized upskilling of the model depending on a mistake pattern of the user.

### Claim 5 (depends on 1)

The training method as claimed in claim 1, wherein the training module comprises a user mistake mining module that detects mistake patterns in users, which are mined without referring to the skilled strategy mode.

### Claim 6 (depends on 1)

The training method as claimed in claim 1, wherein the training module comprises a skill scoring module that generates a skill score for the user depending on a deviation of the user decision from the skilled strategy model, wherein this skilled strategy model is used as a benchmark for generating the skill score for the user.

### Claim 7 (depends on 1)

The training method as claimed in claim 1, wherein the training module comprises a skill-based module that conducts skill-based campaigns and services that are dependent on a skill of the user that is obtained from a skill score.

### Claim 8

A training system for training a user performing a predefined operation, to make a best decision based on previous best decisions made by skilled users in the predefined operation, the training system comprising:
at least one processor coupled with a first storage module and a second storage module;
the first storage module to store a first log of the decisions made by the skilled users during the predefined operation;
a first pre-processing module in communication with the first storage module to pre-process the first log to generate a first multi-dimensional image array of each state of the predefined operation;
a training module to train a model based on the first log at a predetermined state of the predefined operation, which is captured as the first multi-dimensional image array to generate a skilled strategy model;
the second storage module to store a second log of decisions made by the user in the predefined operation;
a second pre-processing module in communication with the second storage module to pre-process the second log to generate a second multi-dimensional image array; and
a comparison module to compare the second multi-dimensional image array with the skilled strategy model to generate a prediction of the best decision to be made by the user.

### Claim 9 (depends on 8)

The training system as claimed in claim 8, wherein the first pre-processing module generates snapshots based on the pre-processed first log to make the first log machine readable and machine learnable.

### Claim 10 (depends on 8)

The training system as claimed in claim 8, wherein the prediction of best move is displayed on a user device being used by the user so that the user is prompted to execute the best move.

### Claim 11 (depends on 8)

The training system as claimed in claim 8, wherein the training module comprises a personalized upskilling module that performs personalized upskilling of the model depending on a mistake pattern of the user.

### Claim 12 (depends on 8)

The training system as claimed in claim 8, wherein the training module comprises a user mistake mining module that detects mistake patterns in users, which are mined without referring to the skilled strategy model.

### Claim 13 (depends on 8)

The training system as claimed in claim 8, wherein the training module comprises a skill scoring module that generates a skill score for the user depending on a deviation of the user decision from the skilled strategy model, wherein this skilled strategy model is used as a benchmark for generating the skill score for the user.

### Claim 14 (depends on 8)

The training system as claimed in claim 8, wherein the training module comprises a skill-based module assigns skill-based campaigns and services that are dependent on a skill of the user that is obtained from a skill score.

### Claim 15

A non-transitory computer program product to train a user in performing a predefined operation to make a best decision based on previous decisions made by skilled users in the predefined operation, when executed by a computer, the computer program product comprising programmed codes to:
process stored instructions in a first storage module and a second storage module via at least one processor coupled with the first storage module and the second storage module;
store, via the first storage module, a first log of the decisions made by the skilled users during the predefined operation;
pre-process, via a first pre-processing module in communication with the first storage module, the first log to generate a first multi-dimensional image array of each state of the predefined operation;
train, via a training module, a model based on the first log at a predetermined state of the predefined operation that is captured as the first multi-dimensional image array to generate a skilled strategy model;
store, via the second storage module, a second log of decisions made by the user in the predefined operation;
pre-process, via a second pre-processing module in communication with the second storage module, the second log to generate a second multi-dimensional image array; and
compare, via a comparison module, the second multi-dimensional image array with the skilled strategy model to generate a prediction of the best decision to be made by the user.
