# US 2026/0087027 A1 · Method, kitchen appliance, service device and system for generating a culinary knowledge base

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

- Number: US 2026/0087027 A1
- Document ID: US20260087027A1
- Status: Published application
- Kind: A1
- Application number: 19/330,895
- Filed: 2025-09-17
- Published: 2026-03-26
- Earliest claimed: 2024-09-25
- Assignees: BSH Hausgeräte GmbH
- Applicants: BSH Hausgeräte GmbH
- Inventors: Gioninta NAKOU-FRANKE; Yi Heng CHEONG; Mohit KULKARNI; Kaushal SAGAR
- Web page: https://patentgrep.com/patent/US20260087027A1

## CPC

- G06F 16/25: Integrating or interfacing systems involving database management systems
- G06Q 50/12: Hotels or restaurants
- A47J 36/321: {the electronic control being performed over a network, e.g. by means of a handheld device}
- A47J 44/00: Multi-purpose machines for preparing food {with several driving units}
- G06F 40/30: Semantic analysis
- G06N 3/02: Neural networks
- G16H 20/60: relating to nutrition control, e.g. diets
- G09B 19/0092: {Nutrition}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- 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
- A47J: KITCHEN EQUIPMENT; COFFEE MILLS; SPICE MILLS; APPARATUS FOR MAKING BEVERAGES
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- 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)

## Abstract

A method for generating a culinary knowledge base includes determining user actions during preparation of a dish and determining (205, 215) a first cooking recipe. The first cooking recipe includes preparation steps which are similar to the user actions. A second cooking recipe is generated. The second cooking recipe includes preparation steps that match the user actions for preparing the dish. The second cooking recipe is stored as a variant of the first cooking recipe. A kitchen appliance, a service device and a system for generating a culinary knowledge base, are also provided.

## Claims

### Claim 1

A method for generating a culinary knowledge base, the method comprising:
determining user actions during preparation of a dish;
determining a first cooking recipe, the first cooking recipe including preparation steps being similar to the user actions;
generating a second cooking recipe, the second cooking recipe including preparation steps matching the user actions for preparing the dish; and
storing the second cooking recipe as a variant of the first cooking recipe.

### Claim 2 (depends on 1)

The method according to claim 1, which further comprises determining that a dissimilarity between the first and second cooking recipes is lower than a first predetermined threshold.

### Claim 3 (depends on 1)

The method according to claim 1, which further comprises determining that a dissimilarity between the first and second cooking recipe is greater than a second predetermined threshold.

### Claim 4 (depends on 2)

The method according to claim 2, which further comprises associating each of the first and second cooking recipes with a respective point in a latent space of an artificial neural network, and determining a dissimilarity between the first and second cooking recipes based on a distance between the associated points in the latent space.

### Claim 5 (depends on 3)

The method according to claim 3, which further comprises associating each of the first and second cooking recipes with a respective point in a latent space of an artificial neural network, and determining a dissimilarity between the first and second cooking recipes based on a distance between the associated points in the latent space.

### Claim 6 (depends on 1)

The method according to claim 1, which further comprises selecting the first cooking recipe from a repository including a plurality of cooking recipes.

### Claim 7 (depends on 1)

The method according to claim 1, which further comprises generating the first cooking recipe in text form using a Large Language Model.

### Claim 8 (depends on 1)

The method according to claim 1, which further comprises providing an ingredient or intermediate product for the dish or an amount or state of the ingredient or intermediate product, as part of the user actions.

### Claim 9 (depends on 1)

The method according to claim 1, which further comprises providing a processing of an ingredient or intermediate product as part of the user actions.

### Claim 10 (depends on 1)

The method according to claim 1, which further comprises providing a use of one or more kitchen appliances as part of the user actions.

### Claim 11 (depends on 1)

The method according to claim 1, which further comprises storing the second cooking recipe only when the user expresses contentment with the prepared dish.

### Claim 12 (depends on 1)

The method according to claim 1, which further comprises storing the second cooking recipe with a reference to characteristics of the user.

### Claim 13 (depends on 11)

The method according to claim 11, which further comprises providing a dissimilarity of associated user characteristics as part of a dissimilarity between cooking recipes.

### Claim 14 (depends on 13)

The method according to claim 13, which further comprises determining that a dissimilarity between the first and second cooking recipes is lower than a first predetermined threshold, and providing the dissimilarity of associated user characteristics as part of the dissimilarity between the first and second cooking recipes.

### Claim 15 (depends on 13)

The method according to claim 13, which further comprises determining that a dissimilarity between the first and second cooking recipe is greater than a second predetermined threshold, and providing the dissimilarity of associated user characteristics as part of the dissimilarity between the first and second cooking recipes.

### Claim 16

A kitchen appliance, comprising:
a sensor for determining a user action during preparation of a dish; and a
processing device adapted:
to determine, based on determined user actions, a first cooking recipe, the first cooking recipe including preparation steps being similar to the user actions;
to generate a second cooking recipe, the second cooking recipe including preparation steps matching the user actions for preparing the dish; and
to store the second cooking recipe as a variant of the first cooking recipe.

### Claim 17

A service device, comprising means adapted to determine a dissimilarity between two cooking recipes based on an artificial neural network.

### Claim 18 (depends on 16)

A system for generating a culinary knowledge base, the system comprising:
at least one kitchen appliance according to claim 16;
a service device adapted to determine a dissimilarity between two cooking recipes based on an artificial neural network; and
a storage device for a plurality of cooking recipes.
