Food Recommender System
Chirag Vithalani
May 2, 2017
“Eat food. Not too much. Mostly plants. That is the short answer to the supposedly, incredibly complicated and confusing question of what we humans should eat to be maximally healthy.” – In Defense of Food: An Eater’s Manifesto Book by Michael Pollan.
There are two ways to implement food recommender system.
1.Recommend Recipes: This would use similarity procedures to recommend recipes which are most liked by users and in turn which one is healthier option.
2.Recommend Food(items): The second type of food recommender would suggest which foods to replace with which healthier option. A system which analyses shopping receipts (or barcodes) and then recommends healthier food choices supplying the nutrients missing in the user’s diet. Unlike recipe recommender systems, food recommender systems wouldn’t take user likes and dislikes into account. For example, https://play.google.com/store/apps/details?id=com.fooducate.nutritionapp which is app where the user can scan the barcode of the food item and app can grade that food item by analyzing ingredients.
Challenges for Recommending Recipes
- To get the dietary requirement of the user.
- To get user ratings of specific recipes?
- How can the user enter information about the one’s last meals? and how to keep user motivated such that one keeps entering such data. People tend to forget what they have eaten or how much they have eaten.
- And there is always Cold-start problem
Challenges for Recommending food items
- How to get information about food additives and how to know which product is more healthy than other?
- How to create the database where we can get food additives present in the product?
Possible solution for recommending healthy food items
All food additive list is given by :https://en.wikipedia.org/wiki/International_Numbering_System_for_Food_Additives ( which is mostly based on E-Numbering system of European Union ). The link above shows which food additives are approved in which countries.
Get harmful additive list from here : http://mphprogramslist.com/50-jawdroppingly-toxic-food-additives-to-avoid/
When the user scans the barcode; if information of additives not present then ask the user to take picture of it ( which shows the list of additives ) and that way build such database.