A Large-scale Recipe and Meal Data Collection as Infrastructure for Food Research

A Large-scale Recipe and Meal Data Collection as Infrastructure for Food Research Jun Harashima, Michiaki Ariga, Kenta Murata, Masayuki Ioki Cookpad Inc. Yebisu Garden Place Tower 12F, 4-20-3 Ebisu, Shibuya-ku, Tokyo, 150-6012, Japan {jun-harashima, michiaki-ariga, kenta-murata, masayuki-ioki}@cookpad.com Abstract Everyday meals are an important part of our daily lives and, currently, there are many Internet sites that help us plan these meals. Allied to the growth in the amount of food data such as recipes available on the Internet is an increase in the number of studies on these data, such as recipe analysis and recipe search. However, there are few publicly available resources for food research; those that do exist do not include a wide range of food data or any meal data (that is, likely combinations of recipes). In this study, we construct a large-scale recipe and meal data collection as the underlying infrastructure to promote food research. Our corpus consists of approximately 1.7 million recipes and 36000 meals in cookpad, one of the largest recipe sites in the world. We made the corpus available to researchers in February 2015 and as of February 2016, 82 research groups at 56 universities have made use of it to enhance their studies. Keywords: recipe, meal, food research 1. Introduction Nowadays many people plan meals using the Internet. Be- cause meals need to be planned several times a day, this is a time-consuming and labor-intensive task. However, ow- ing to the many Internet sites available to help users find recipes and plan their meals (e.g., cookpad in Japanese,1 and Yummly in English2), such planning has become much easier. Alongside the increase in the amount of available food data such as recipes, there has been a corresponding in- crease in the number of studies on these data such as recipe analysis (Kiddon et al., 2015; Jermsurawong and Habash, 2015; Maeta et al., 2015; Sasada et al., 2015) and recipe search (Yasukawa et al., 2014; Wang et al., 2008). Further- more, many conferences and workshops have recently been held to promote research on these topics. Nevertheless, there are few publicly available resources for food research. Additionally, such resources are not widely applicable as they do not include a wide range of food data (a few dozen − hundreds of thousands of recipes) or any meal data (data on likely combinations of recipes). Thus, a novel resource is required to promote food research. In this study, we constructed a large-scale recipe and meal data collection as the underlying infrastructure for food re- search. We organized approximately 1.7 million recipes, 36000 meals, and their related data (e.g., reviews) in cook- pad and made the collected data available to researchers. In this paper, we report on how we designed our corpus and how this has contributed to enhancing food research. The remainder of this paper is organized as follows. Sec- tion 2. briefly describes relevant previous studies. Sec- tions 3. and 4. discuss, respectively, the recipe and meal data in our corpus. Section 5. summarizes the contribu- …

Content truncated to 500 words