Learning Personal Food Preferences via Food Logs Embedding

TL;DR

This paper presents a method to learn individual food preferences from food logs using embeddings, improving personalized diet recommendations for managing chronic diseases like diabetes.

Contribution

It introduces a novel approach to extract food preferences from noisy food logs using embedding techniques and provides new metrics for evaluation.

Findings

Identifies 82% of a user's top 10 foods.

Uses word embeddings to classify food categories.

Method is publicly available for use and further research.

Abstract

Diet management is key to managing chronic diseases such as diabetes. Automated food recommender systems may be able to assist by providing meal recommendations that conform to a user's nutrition goals and food preferences. Current recommendation systems suffer from a lack of accuracy that is in part due to a lack of knowledge of food preferences, namely foods users like to and are able to eat frequently. In this work, we propose a method for learning food preferences from food logs, a comprehensive but noisy source of information about users' dietary habits. We also introduce accompanying metrics. The method generates and compares word embeddings to identify the parent food category of each food entry and then calculates the most popular. Our proposed approach identifies 82% of a user's ten most frequently eaten foods. Our method is publicly available on (https://github.com/aametwally/LearningFoodPreferences)