---
license: afl-3.0
task_categories:
  - text-generation
language:
  - en
tags:
  - recipes
  - food
  - diabetic
pretty_name: DUT Diabetic Friendly Recipes
size_categories:
  - n<1K
---

# Durban University of Technology - Diabetic Friendly Recipes

![image/png](assets/dut.png)

## Description

This dataset features a collection of recipes that prioritise the use of low and medium glycemic index (GI) ingredients to create flavourful and nutritious dishes aimed at reducing the risk of type-2 diabetes.
In a collaborative effort between the Food and Nutrition academics and students at the Durban University of Technology, along with a Senior Developer, a database of thoroughly tested, high-quality diabetic-friendly recipes was used to train a model capable of generating reliable, diabetic-friendly recipes.
During model training, unfamiliar recipes were tested by the Department of Food and Nutrition.

### Recipe Components
- Ingredients in household measurements
- Method: step-by-step preparation method

## Limitations

Cultural sensitivity:  Whilst the dataset of recipes includes a variety of global recipes, it may not account for some cultural food practices, ingredients, or preferences.

Missing ingredients: Not all available ingredients in the global food system have been accounted for in the dataset.

Serving size: The dataset only provides a rough estimate of the number of servings per recipe.


## Citation Information

```
Prof Ashika Naicker*,  Mr Shaylin Chetty,  Ms Riashnie Thaver*, Ms. Anjellah Reddy*, Dr. Evonne Shanita Singh*, Dr. Imana Pal*, Dr. Lisebo Mothepu*.

*Durban University of Technology, Faculty of Applied Sciences, Department of Food and Nutrition,  Durban, South Africa
```