Machine Learning based Omnifood: Nutritional Meal Drop

There are many portable applications for weight and diet management available these days. Furthermore, there are few examples of food recommendation systems that provide nutritional information about reasonable food choices to the user. An enormous number of people are extraordinarily busy with their positions, loved ones, and other basic exercises, which doesn't take into account cooking. This could lead a poor diet routine and lasting health consequences. It is highly required to solve this issue by using a web-based methodology. Clients can use the developed web page to pick their eating routine and food sources they like and dislike, and our computation will make a custom and individual step by step weekly plan. The proposed web application guarantees the client to obtain all of the required enhancements and supplements, notwithstanding the eating routine. The proposed application works with the eateries and other cooking accomplices to truly get ready and convey all dinners from the delivered feast plans to chosen regions. This paper proposes Multinomial Naive Bayes and this proposed algorithm, classified for food reviews. The proposed research work has created a webpage using HTML5, CSS3, and JavaScript and classified food reviews using machine learning. Four different algorithms are executed on the proposed dataset, they are Bernoulli Naive Bayes, Gaussian Naive Bayes, Logistic Regression and Multinomial Naive Bayes. In algorithm evaluation, a performance matrix has been formed i.e., precision, recall and accuracy of Multinomial Naive Bayes are observed to be high when compared to other algorithms. So, the Multinomial Naive Bayes algorithm is utilized in the proposed model.