Abstract:
This report presents a comprehensive study that combines large-scale Exploratory Data Analysis (EDA) with the development of a modern web application, collectively forming the foundation for RecipeDB — an intelligent and interactive recipe discovery platform. The dataset under consideration comprises 584,572 diverse recipe records, each annotated with 21 attributes includ- ing ingredients, preparation time, cooking instructions, nutrition facts, categories, and cuisine types. The EDA component focused on identifying patterns in ingredient usage, evaluating the dis- tribution of preparation and cooking times, detecting data inconsistencies, and constructing a robust dietary tagging framework encompassing classifications such as Jain, Vegetarian, Egg- based, and Non-Vegetarian. This process also involved significant data cleaning, normalization, and the detection of duplicate or semantically similar entries, ensuring high data quality and consistency for downstream applications. The insights gained from the analysis were used to inform the design of a web-based interface, which allows users to search, filter, and explore recipes using a wide variety of parameters includ- ing ingredients, nutrients, utensils, and dietary preferences. The frontend of the application was developed using React (Vite) and Tailwind CSS, while the backend was built using Node.js with Express and connected to a MySQL database. Key features include a responsive and visually engaging UI, real-time search suggestions, nutrient-based sliders, and dark/light mode toggling for enhanced usability. The project demonstrates the integration of data science and software engineering practices to build a scalable and user-centric platform. It not only showcases how rich insights can be derived from raw recipe data but also how these insights can be operationalized through a well-designed application to improve user interaction and food information accessibility.
