HYBRID EVENT: You can participate in person at Rome, Italy or Virtually from your home or work.
HYBRID EVENT
September 16-18, 2024 | Rome, Italy
FAT 2024

Vidip Chabbra

Vidip Chabbra, Speaker at Food Science Conferences
PES University, India
Title : Intelligent ingredient management for recipes

Abstract:

The culinary landscape is characterized by a rich diversity of recipes, each comprising a unique combination of ingredients. However, the inconvenience of missing ingredients often hinders the cooking process, prompting the need for intelligent systems that can seamlessly recommend substitutions without compromising the flavour profile of the dish. In response to this challenge, our research endeavors to develop an advanced system for Intelligent Ingredient Management in Recipes.

The project begins with the extraction of comprehensive recipe, ingredient, and nutrient data from the Spoonacular API, a vast repository of culinary information. Subsequent data cleaning processes involve the removal of stopwords from both recipes and ingredients, creating a refined dataset conducive to advanced analysis. To enhance the model's understanding of recipes, encodings are applied, where binary representations signify the presence or absence of specific meal categories (e.g., breakfast, lunch, dinner).

A critical component of our methodology involves the construction of a bipartite graph that establishes connections between recipes and ingredients. This graph facilitates the identification of neighboring ingredients within the context of specific recipes, forming the basis for effective substitutions. The integration of various natural language processing techniques further refines the model's ability to discern semantic similarities between ingredients.

The arsenal of techniques employed for ingredient substitution includes word2vec, WordNet, cosine similarity, count vectorizer, and the BERT model. These methodologies collectively contribute to the creation of a versatile and adaptive system capable of providing precise and contextually relevant ingredient replacements.

The paper delineates the intricacies of our "ingredient substitution" function, which takes the bipartite graph and specific recipe information as input to recommend suitable alternatives for missing ingredients. This function serves as the linchpin in our intelligent ingredient management system, embodying the culmination of our research efforts.

Through this research, we aim to pave the way for enhanced culinary experiences by empowering individuals to seamlessly adapt recipes to ingredient availability while preserving the essence of each dish. The subsequent sections delve into the methodologies employed, the results obtained, and the implications of our findings for the broader field of recipe management and intelligent systems.

Audience Take Away:
• Usefulness for the Audience: The audience, comprising cooking enthusiasts, chefs, and individuals interested in dietary restrictions, will learn about ingredient substitutions, enabling them to adapt recipes according to dietary needs, food allergies, or ingredient availability. They'll gain insights into how to replace ingredients without compromising taste or quality.
• Job Relevance: For chefs and culinary professionals, this knowledge can significantly impact their job performance. It allows them to cater to a broader range of dietary requirements, enhancing customer satisfaction. Moreover, it equips them with the skills to adapt recipes without compromising the essence of the dish, improving their versatility and resourcefulness in the kitchen.
• Research Expansion for Faculty: This project could serve as a valuable resource for other faculty members interested in culinary arts, nutrition, or food science. It may provide a foundation for further research into the science behind ingredient replacements, offering insights into how different ingredients affect taste, texture, and nutritional value in recipes.
• Practical Solution for Designers: It provides a practical solution by simplifying the adaptation of recipes. Designers, in this case, referring to culinary recipe creators or developers, can use this information to efficiently modify recipes for specific dietary needs or ingredient availability without compromising the quality of the final dish.
• Improvement in Design Accuracy: Understanding ingredient replacements can significantly enhance the accuracy of recipe design. It enables creators to offer alternative versions of recipes that are tailored to different dietary preferences or restrictions, expanding the reach and relevance of their culinary creations.

Biography:

Vidip Chabbra studies Computer Science Engineering at PES University, India and will graduate in as undergraduate in 2024. He is currently doing his internship in Livesitter as AI Intern. During his college years he was the club head of PredictThis the ML club. He received scholarship every semester for distinction.

Watsapp