Abstract:
In today's fast-paced world, ordering food is often seen as a convenient and
time-efficient option. However, cooking at home is a significantly healthier choice
than ordering food. Despite the abundance of recipes available in cookbooks and
on the internet, selecting a suitable recipe may be time-consuming task. To alleviate
this issue, the author proposes the development of an Android-based cookbook that
employs a recommendation system to assist users in discovering appropriate
recipes.
The developed application provides recipe recommendations based on the
similarity of recipe categories, titles, and ingredients owned by the user. By utilizing
a word to represent the title of recipe and ingredients as the user's input within the
application, the system receives the user's input and employs a content-based
approach and mathematical equations to generate personalized recipe
recommendations. The front-end of the application is built with the Flutter
framework, while the back-end employs the Python framework for
recommendation model creation and implementation. The application
implementation utilizes a dataset consisting of over 7000 recipes sourced from
Kaggle.com
