Live Personalized Nutrition Recommendation Engine

Abstract

Dietary choices are the primary determinants of prominent dis- eases such as diabetes, heart disease, and obesity. Human health care providers, such as dietitians, cannot be at the side of every user at all times to manually guide them towards optimal choices. Automated adaptive guidance fused with expert knowledge can use multimedia data to technologically scale health guidance without human intervention. Addressing the correct granularity of recommendations (in this case meal dishes) is essential for effortless decision making. Thus we make a decision support system using multi-modal data relying on timely, contextually aware, personalized data to find local restaurant dishes to satisfy a user’s needs. Algorithms in this system take nutritional facts regarding products, efficiently calculate which items are healthiest, then re-rank and filter results to users based on their personalized health data streams and environmental context. Our recommendation engine is driven by the primary goal of lowering the barriers to a personalized healthy choice when eating out, by distilling dish suggestions to a single contextually aware and easily understood score.

Food Recommendation: Framework, Existing Solutions, and Challenges

“…As representative work, Nag et al [8] proposed a live personalized nutrition recommendation system. As shown in used these sensors to calculate a live estimate of the user's daily nutritional requirements.…”

“…In addition, we can even associate visual information and other modality information with this KG to build multimodal food knowledge graph. 8 https://en.wikipedia.org/wiki/Lists of foods…”

“…Existing multimedia research has made great progress in improving the recommendation performance and experience in many fields such as movies and POI, yet largely lags in the food domain. To the best of our knowledge, although relevant works on food recommendation [7], [8], [9] have received more attention in the multimedia community, there are very few systematic reviews, which provide a unified framework and comprehensive summary of current efforts in food recommendation. Because of huge potentials in food recommendation, the time has come for the multimedia field to give a survey on food recommendation, which can help researchers from relevant communities better understand the strength and weakness of existing methods.…”

Continuous Health Interface Event Retrieval

“…The details for this project could be found in [32]. A nutrition guidance/recommendation system could utilize such a system to calculate an individual's electrolyte requirements by combining their location stream with local weather to estimate their activity levels and local temperature, which in turn allows us to estimate how much sodium they lost through sweat [34]. This can be generalized to other macronutrients too.…”

A Food Recommender System Considering Nutritional Information and User Preferences

“…Here, there have been some researches focused on restaurant menu recommendation such as Ntalaperas et al [37], focused on ranking dishes based on medical conditions, users' settings and preferences based on past rankings, but specifically focused on a restaurant menu. In a different direction, we detect a small group of research works focused on processing multimodal data, such as Nag et al [35] propose a live personalized nutrition recommendation engine that uses multimodal contextual data including GPS location, barometer, and pedometer output to calculate a live estimate of the user's daily nutritional requirements, that are then used to rank the meals based on how well they fulfill the individual's nutritional needs. In this direction, Ge et al [19] propose a food recommender system developed on a mobile platform, which not only offers recipe recommendations that suit the user's preference but is also able to take the user's health into account, supported by wearable technologies.…”