Recommender System for Meal Menus using a Potential Model

Takanobu Nakahara, Hiroyuki Morita, and Yukinobu Hamuro


Recommender system, Meal menu data, Data mining


Recommender systems have become more important in terms of e-business. The rapid increase of information provided to customers has resulted in an information explosion.As a result, we have to provide customers with information that is really necessary to them from the huge information resources available. E-commerce firms such as Amazon and Netflix have successfully used recommender systems to increase sales and improve customer loyalty.Some systems may be useful in another aspect: home meal menus and learning about cooking techniques are themes in which people who cook at home take a considerable interest.In this study, we propose a recommender system that incorporates individual preference as a potential model and unpredictability. By means of computational experiments using practical data, we have compared our method with others. From the results, we show that our method has good performance for the data used.

Important Links:

Go Back