Eating healthier: Exploring nutrition information for healthier recipe recommendation

Abstract

With the booming of personalized recipe sharing networks (e.g., Yummly), a deluge of recipes from different cuisines could be obtained easily. In this paper, we aim to solve a problem which many home-cooks encounter when searching for recipes online. Namely, finding recipes which best fit a handy set of ingredients while at the same time follow healthy eating guidelines. This task is especially difficult since the lions share of online recipes have been shown to be unhealthy.In this paper we propose a novel framework named NutRec, which models the interactions between ingredients and their proportions within recipes for the purpose of offering healthy recommendation. Specifically, NutRec consists of three main components: 1) using an embedding-based ingredient predictor to predict the relevant ingredients with user-defined initial ingredients, 2) predicting the amounts of the relevant ingredients with a multi-layer perceptronbased network, 3) creating a healthy pseudo-recipe with a list of ingredients and their amounts according to the nutritional information and recommending the top similar recipes with the pseudo-recipe. We conduct the experiments on two recipe datasets, including Allrecipes with 36,429 recipes and Yummly with 89,413 recipes, respectively. The empirical results support the framework's * This is to indicate the corresponding authors.intuition and showcase its ability to retrieve healthier recipes.

Personal Information Disclosure via Voice Assistants: The Personalization–Privacy Paradox

“…To make the VA services more 'personalized', collecting and analyzing personal information is required for tailoring the services and related contents as per the needs of the target users. In general, the users are willing to share their personal information in return of personalized services, like getting greetings on their birthdays with special discount coupons, book recommendations, travel suggestions or even recipe suggestions based on their health and nutritional requirements [49][50][51][52]. Another research shows that the users are very positive about receiving personalized services and are even willing to watch advertisements in return for a specific product they had been looking for [53].…”

“…As stated previously, the privacy calculus framework enables the users to do a risk-benefit analysis of the motivational factors that either act as a positive or negative catalyst of personal information disclosure. If the users feel that they get some benefits like greater personalized services [49][50][51][52][53][54][55], enjoyment [4,[56][57][58][59][60], convenience [81] or monetary rewards by disclosing their private information [82], then they will forgo some level of their privacy in return for the perceived benefits. Therefore, it is the subjective evaluation of the potential gains or the positives.…”

Effects and challenges of using a nutrition assistance system: results of a long-term mixed-method study

“…Approaches to health-focused food recommender systems have used energy balancing (Ge et al 2015b), distance from an estimated nutritional requirements (Elsweiler et al 2015b), or re-weighting according to health metrics (Trattner and Elsweiler 2017b). Beyond prediction of ratings or ranking based on an existing set of recipes, other efforts have been focused on suggesting healthier ingredient substitutes (Achananuparp and Weber 2016) or on generating healthier pseudo recipes (Chen et al 2019). The implemented algorithm in the Nutrilize system is a content-based approach for both health and taste estimation that integrates both energy balancing and nutritional requirements.…”

Semantic-aware transformation of short texts using word embeddings: An application in the Food Computing domain

“…It has been mentioned the many advantages of embedding models referring to fusion heterogeneous food data for multiple purposes, where nutritional and social media textual data are integrated (Salvador et al, 2017) more specialized in resolving image recognition tasks rather than language processing. In (Chen et al, 2019), the authors used a word embedding model to detect ingredient relations to create pseudo-recipes. They used a model trained on a list of recipes to detect which ingredients appear together in recipes.…”