Fuzzy Recipe Recommender is an expert system developed using Fuzzy Jess Toolkit, which could be used to receive recommendations about different recipes based on the vague, imprecise, uncertain, ambiguous, inexact, or probabilistic nature of quantity of nutrients (Proteins, Carbohydrates & Fats) that the user wants to consume. This is just a prototype so it has 24 test cases right now. However, the system can be scaled to work for a large set of recipes. In this expert system, I am categorizing the Recipe Recommendation in 3 categories:
- Main Course.
The system first takes a keyword from the above mentioned three categories as an input, which will allow the user to choose what kind of recipe he/she wants to cook in broader perspective.
Next, the system asks user about the quantity of proteins, carbohydrates and fats that the user wants to consume and based on his/her input an appropriate name of recipe and a link to its YouTube or textual method of preparation with ingredients is displayed as output.
The following are the fuzzy terms that I have used to fuzzify the system and these could be chosen for the following nutrients: Proteins – low/high Carbohydrates - less/more Fats - less/more