IPFC: machine learning based intelligent prediction of food preference for cooking

In recent years, China's material standard of life has improved as a result of the country's economic growth, the number of obese people in China has roughly tripled in 15 years, with half of all adults overweight or obese. Obesity is a major independent risk factors for chronic diseases such as cardiovascular disease, diabetes, closely related to the occurrence of such as esophageal cancer development also. In order to deal with this 'public health crisis', many ingredient recommendation systems have been developed for cooking healthy diets. In order to improve their performance, this study presents IPFC, an innovative and intelligent prediction tool, to predict people's food preference for cooking, where they can give more targeted recommendations. Our strategy is based on a dataset gathered through a real-world questionnaire, in which the questions are about participants' eating and behavior habits and their basic information. We utilize data mining methods on such a dataset. During the assessment tests, machine learning models such as ANN, DT, RF, SVM, GBDT, LR, and NB are evaluated for their ability to classify the dataset into several classes. The experimental findings demonstrate that SVM achieves a relatively higher accuracy than other considered machine learning models, whici is 50.00% using the test set that consists of a small number of data collected by us.