# US 2021/0265036 A1 · Recipe generation based on neural network

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

- Number: US 2021/0265036 A1
- Document ID: US20210265036A1
- Status: Published application
- Kind: A1
- Application number: 17/318,640
- Filed: 2021-05-12
- Published: 2021-08-26
- Earliest claimed: 2018-11-13
- Assignees: Journey Foods Inc.
- Applicants: Journey Foods Inc.
- Inventors: Riana Lynn; Carol Schmitz
- Web page: https://patentgrep.com/patent/US20210265036A1

## CPC

- G16H 20/60: relating to nutrition control, e.g. diets
- G06N 3/0499: Feedforward networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06Q 30/0282: Rating or review of business operators or products
- G06Q 30/0283: Price estimation or determination
- G06Q 30/0621: {by configuring or customising goods or services}
- G06Q 30/0623: {by investigating goods or services}
- G06Q 30/0631: {Recommending goods or services}
- G06Q 50/12: Hotels or restaurants
- G06Q 50/22: Social work or social welfare, e.g. community support activities or counselling services
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- 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

## Abstract

An illustrative aspect of the present disclosure is a system and method of generating a recipe, which includes determining rankings for a plurality of food ingredients for a target food product according to at least one of a health-based feature, a cost, or a prevalence of the plurality of food ingredients. A neural network can receive, as inputs, the rankings of the plurality of food ingredients, and a plurality of existing recipes for food products in a same category as the target food product. The neural network can generate, as output, according to the rankings of the plurality of food ingredients and the plurality of existing recipes, a new recipe for the target food product that incorporates a select food ingredient from the plurality of food ingredients.

## Claims

### Claim 1

A method of generating a recipe, comprising:
determining rankings for a plurality of food ingredients for a target food product according to at least one of: a health-based feature, a nutritional property or measure, a cost, or a prevalence of the plurality of food ingredients;
receiving, by a neural network as inputs, the rankings of the plurality of food ingredients, and a plurality of existing recipes for food products in a same category as the target food product; and
generating, by the neural network as output, according to the rankings of the plurality of food ingredients and the plurality of existing recipes, a new recipe for the target food product that incorporates a select food ingredient from the plurality of food ingredients.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
receiving, by the neural network as another input, properties of each of the plurality of food ingredients; and
generating, by the neural network, according to the rankings of the plurality of food ingredients, the plurality of existing recipes and the properties of each of the plurality of food ingredients, the new recipe.

### Claim 3 (depends on 1)

The method of claim 1, wherein the new recipe has a higher probability of incorporating a first food ingredient than a second food ingredient with a ranking that is lower than that of the first food ingredient in the plurality of food ingredients, as the select food ingredient.

### Claim 4 (depends on 3)

The method of claim 3, wherein the select food ingredient has a ranking that is lower than that of at least one food ingredient in the plurality of food ingredients.

### Claim 5 (depends on 1)

The method of claim 1, further comprising tagging a first recipe of the plurality of existing recipes, to limit a type of modification to the first recipe by the neural network in generating the new recipe.

### Claim 6 (depends on 1)

The method of claim 1, wherein the target food product comprising at least one of: a snack, a baked food, a confectionery, a synthetic plant-based product, tofu product, meat or diary alternative, an energy food item, a beverage, a meal substitute, a dietary supplement, a superfood, fresh food product, a shelf-stable food, a frozen or refrigerated food, a meat product, or a dairy product.

### Claim 7 (depends on 1)

The method of claim 1, further comprising generating, by the neural network, the new recipe for the target food product that incorporates multiple food ingredients from the plurality of food ingredients.

### Claim 8 (depends on 1)

The method of claim 1, further comprising generating, by a label engine, a nutritional description for the new recipe.

### Claim 9 (depends on 1)

The method of claim 1, wherein determining the rankings comprises:
determining first rankings for the plurality of food ingredients according to at least one health-based feature of each of the plurality of food ingredients;
determining second rankings for the plurality of food ingredients according to a cost of each of the plurality of food ingredients;
determining third rankings for the plurality of food ingredients according to a prevalence of each of the plurality of food ingredients; and
determining the rankings for the plurality of food ingredients according to the first rankings, the second rankings and the third rankings.

### Claim 10 (depends on 1)

The method of claim 1, wherein the health-based feature of the plurality of food ingredients includes a nutritional measure of one of the plurality of food ingredients.

### Claim 11

A system for generating a recipe, comprising:
a storage device configured to store a database of a plurality of food ingredients for a target food product;
at least one processor configured to determine rankings for the plurality of food ingredients for the target food product according to at least one of: a health-based feature, a nutritional property or measure, a cost, or a prevalence of the plurality of food ingredients; and
a neural network implemented by the at least one processor, configured to:
receive as inputs the rankings of the plurality of food ingredients, and a plurality of existing recipes for food products in a same category as the target food product; and
generate as output, according to the rankings of the plurality of food ingredients and the plurality of existing recipes, a new recipe for the target food product that incorporates a select food ingredient from the plurality of food ingredients.

### Claim 12 (depends on 11)

The system of claim 11, wherein the neural network is configured to:
receive properties of each of the plurality of food ingredients; and
generate, according to the rankings of the plurality of food ingredients, the plurality of existing recipes and the properties of each of the plurality of food ingredients, the new recipe.

### Claim 13 (depends on 11)

The system of claim 11, wherein the new recipe has a higher probability of incorporating a first food ingredient than a second food ingredient with a ranking that is lower than that of the first food ingredient in the plurality of food ingredients, as the select food ingredient.

### Claim 14 (depends on 13)

The system of claim 13, wherein the select food ingredient has a ranking that is lower than that of at least one food ingredient in the plurality of food ingredients.

### Claim 15 (depends on 11)

The system of claim 11, wherein the at least one processor is configured to tag a first recipe of the plurality of existing recipes, to limit a type of modification to the first recipe by the neural network in generating the new recipe.

### Claim 16 (depends on 11)

The system of claim 11, wherein the target food product comprises at least one of: a snack, a baked food, a confectionery, a synthetic plant-based product, tofu product, meat or diary alternative, an energy bar, a beverage, a meal substitute, a dietary supplement, a superfood, fresh food product, a shelf-stable food, a frozen or refrigerated food, a meat product, or a dairy product.

### Claim 17 (depends on 11)

The system of claim 11, wherein the neural network is configured to generate the new recipe to incorporate multiple food ingredients from the plurality of food ingredients.

### Claim 18 (depends on 11)

The system of claim 11, further comprising a label engine implemented by the at least one processor, configured to generate a nutritional description for the new recipe.

### Claim 19 (depends on 11)

The system of claim 11, the at least one processor is configured to:
determine first rankings for the plurality of food ingredients according to at least one health-based feature of each of the plurality of food ingredients;
determine second rankings for the plurality of food ingredients according to a cost of each of the plurality of food ingredients;
determine third rankings for the plurality of food ingredients according to a prevalence of each of the plurality of food ingredients; and
determine the rankings for the plurality of food ingredients according to the first rankings, the second rankings and the third rankings.

### Claim 20 (depends on 11)

The system of claim 11, wherein the health-based feature of the plurality of food ingredients includes a nutritional measure of one of the plurality of food ingredients.
