Healthy Eating Habits Support System

What will you eat today?

When you decide on your daily menu, what do you consider? While many people want to eat something delicious, it seems likely that some want to eat healthy foods.

Given these factors, deciding what to eat can be a burden for those who cook. When searching for new ideas, people often use recipe books and recipe search sites.

However, published nutrition information is often insufficient, perhaps including only calorie information or information about nutrients in only one recipe. This makes it harder to decide what one must eat to acquire the necessary nutrients.

In addition, when searching recipes, one must select a category or enter keywords. This is not a problem if you have a clear intention such as “I want to eat curry.” However, it becomes less straightforward if you have a vague intention such as “I want to eat something delicious.”

To solve such problems, we have developed a Healthy Eating Habits Support System (HEHSS).

A system that considers nutritional balance and user taste preferences

The HEHSS recommends menus on the basis of nutritional balance matched to user taste preferences. The following figure presents an overview of the HEHSS, which is divided into considerations of nutrients and taste.


Healthy Eating Habits Support System

First, the HEHSS generates many candidate menus on the basis of nutritional balance and stores these in a candidate list, where a menu is defined as a combination of multiple recipes. The HEHSS searches for well-balanced menus by combining a range of recipes from a recipe database.

The HEHSS then picks a menu from the candidate list to match user taste preferences. Here, a Kansei retrieval agent is used to mimic user taste preferences. The Kansei retrieval agent selects a menu in accordance with its own preferences and presents it to the user. The agent learns about user taste preferences on the basis of user evaluations of presented menus.

Implementation

We have implemented a version that can actually be used. The following figure shows the application’s operating screen.

When the application starts, the user enters personal data such as weight and height. Based on this information, the HEHSS calculates the required nutrient amount.


User Resistration

Next, the application goes to the select menu screen. The user first chooses a main dish from recipes presented by the system. Next, the HEHSS presents well-balanced side dishes to accompany the selected main dish. The HEHSS learns user taste preferences through continued use of the application, and the user’s favorite recipes will be displayed at the top of the list.


Main Dish Selection


Side Dishes Selection

Future prospects

Having verified the effectiveness of the system by simulation, we will need to assess the HEHSS’s effectiveness for actual users. In addition, new trends, such as menu proposal based on refrigerator contents and purchased food, invite further investigation.

References

Y.Tokumi, J.Hakamata, M.Tokumaru “Development of a nutritional management system for a healthy eating habits support system”, Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol.17, No.2, pp.324-334, 2013-03.

R.Oku, H.Takenouchi, M.Tokumaru, “Effectiveness of Kansei model in Healthy Eating Habits Support System”, Joint 7th International Conference on Soft Computing and Intelligent Systems and 15th International Symposium on Advanced Intelligent Systems (SCIS & ISIS 2014), pp.335-338, 2014-12 (Kitakyusyu, Japan).

M.Tokumaru, “Implementation of the Healthy Eating Habits Support System Based on User Taste Preferences and Nutritional Balance”, International Conference on Kansei Engineering and Emotion Research 2014, pp.1299-1312, 2014-06 (Linköping, Sweden).

Y.Inagawa, J.Hakamata, M.Tokumaru, “A Support System for Healthy Eating Habits: Optimization of Recipe Retrieval”, the 15th International Conference on Human-Computer Interaction, HCI International 2013 – Posters’ Extended Abstracts, Communications in Computer and Information Science Volume 374, 2013, pp.168-172, 2013-07 (Las Vegas, Nevada, USA).

A.Ishikawa, Y.Tokumi, M.Tokumaru, “Recipe data coding method by using cooking process and quantity of nutrients for healthy eating habits support system”, Joint 6th International Conference on Soft Computing and Intelligent Systems and 13th International Symposium on Advanced Intelligent Systems (SCIS&ISIS2012), pp.197-202, 2012-11 (Kobe, Japan).

Y.Tokumi, J.Hakamata, M.Tokumaru, “Development of a healthy eating habits support system considering user’s taste and health: Generation of menus that are considered nutritionally balanced”, 12th International Symposium on Advanced Intelligent Systems – ISIS2011, pp.167-170, 2011-10 (Suwon, Korea).

J.Hakamata, Y.Tokumi, M.Tokumaru, “Development of a healthy eating habits support system that presents menus considering a user’s taste and health: Optimization of Kansei retrieval system”, 12th International Symposium on Advanced Intelligent Systems – ISIS2011, pp.479-482, 2011-10 (Suwon, Korea).