Machine Learning Model for Predicting the Cuisine Category from a Dish Ingredients

Since ancient times, discussing and recording cooking recipes has been a widespread tradition. The ensuing massive collection of recipes and ingredients has the potential to advance significantly our understanding of both the basics of cooking and food pairing. Many sites have developed cooking suggestion systems or recipe engines in response to the rising popularity of food-based and recipe-sharing. Although this recommendation system provides dish suggestions, it cannot take advantage of the relationship between ingredients and cuisines. Exploring different cultures are achieved by tasting their food. Each country has its own traditional and unique cuisine. The cuisine can be determined by looking at the dish's ingredients. Therefore, the relationship between components and cuisines to predict the recipe cuisine is a motivating task to explore. The problem can be considered a classification problem with several classes. This research project attempts to shift focus away from recipe recommendations and toward investigating and examining the fundamental relationship between the components used in recipes and the various cultures. Support vector machine and associative classification, two basic classification approaches in data science, were used to examine the relationship between different recipes and their ingredient sets. On the dataset sourced from Kaggle, the tests were run. It gave a thorough understanding of other cuisines, ingredient patterns, and the key components of a good recipe. Additionally, the effectiveness of the classifiers used to forecast the cuisines was compared.