Advertisement

Utilizing Various User Moods for Automatic Recipe-Metadata Generation

  • Mayumi UedaEmail author
  • Natsuhiko Takata
  • Yukitoshi Morishita
  • Shinsuke Nakajima
Conference paper

Abstract

In recent times, numerous cooking websites that recommend recipes have been launched. For example, Cookpad [1] and Rakuten Recipe [2] are very popular in Japan. Cookpad contains 2.2 million recipes and 50 million monthly access users, and Rakuten Recipe contains one million recipes. These statistics reflect the high demand for recipe-providing services. We believe that the addition of various metadata to the recipes is effective in improving the accuracy of the recipe recommendation system. For example, if the recipe has metadata such as “good for a bedtime snack”, the system can effectively provide recipes for a specific user purpose, as shown in Fig.1.
Fig. 1

Advantage of the recipe with metadata

Keywords

Feature Vector Gain Score Recommendation Accuracy Master Recipe Cooking Action 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Notes

Acknowledgements

This work was supported in part by the MEXT Grant-in Aid for Scientific Research(C)(#26330351).

References

  1. 1.
  2. 2.
  3. 3.
    Y. Morishit, T. Nakamura, Evaluation of search axis of recipe recommender system based on users’ moods and market needs of food sales functions by using cooking recipe. IEICE Tech. Rep. 112 (75), 79–84 (2012, in Japanese). DE2012-14Google Scholar
  4. 4.
    N. Takata, M. Ueda, Y. Morishita, S. Nakajima, Automatic recipe metadata generating method by considering users’ various moods, in Proceedings of the International Multiconference of Engineers and Computer Scientists 2016, Hong Kong, 16–18 Mar 2016. Lecture Notes in Engineering and Computer Science, pp. 413–418Google Scholar
  5. 5.
    M. Ueda, S. Nakajima, Cooking recipe recommendation method focusing on the relationship between user preference and ingredient quantity, in Transactions on Engineering Technologies, International Multiconference of Engineers and Computer Scientists 2014 (Springer, 2015), pp. 385–395Google Scholar
  6. 6.
    S. Karikome, A. Fujii, A system for supporting dietary habits: planning menus and visualizing nutritional intake balance, in Proceedings of the 4th International Conference on Ubiquitous Information Management and Communication (ICUIMC 2010) (2010), pp. 386–391Google Scholar
  7. 7.
    K. Shirai, H. Ookawa, Constructing a lexicon of actions for the cooking domain toward animation generation. IPSJ Nat. Lang. Process. 2004 (108), 123–128 (2004, in Japanese)Google Scholar
  8. 8.
    Y. Shidochi, I. Ide, T. Takahashi, H. Murase, Finding replaceable materials by cooking recipe mining. IEICE Trans. J94-A (7), 532–535 (2011, in Japanese)Google Scholar
  9. 9.
    T. Ueta, M. Iwakami, T. Ito, A recipe recommendation system based on automatic nutrition information extraction, in Knowledge Science, Engineering and Management. Lecture Notes in Computer Science, vol 7091 (Springer, Berlin/Heidelberg, 2011), pp. 79–90Google Scholar
  10. 10.
    A. Tachibana, S. Wakamiya, H. Nanba, K. Sumiya, Extraction of naming concepts based on modifiers in recipe titles, in The 2014 IAENG International Conference on Internet Computing and Web Services (2014), pp. 507–512Google Scholar
  11. 11.
    MeCab:Yet Another Part-of-Speech and Morphological Analyzer, http://mecab.googlecode.com/svn/trunk/mecab/doc/index.html

Copyright information

© Springer Nature Singapore Pte Ltd. 2017

Authors and Affiliations

  • Mayumi Ueda
    • 1
    Email author
  • Natsuhiko Takata
    • 2
  • Yukitoshi Morishita
    • 3
  • Shinsuke Nakajima
    • 2
  1. 1.University of Marketing and Distribution SciencesNishi-ku, KobeJapan
  2. 2.Kyoto Sangyo UniversityKyotoJapan
  3. 3.Dai Nippon Printing Co.TokyoJapan

Personalised recommendations