Advertisement

Simultaneous Analysis of Subjective and Objective Data Using Coupled Tensor Self-organizing Maps: Wine Aroma Analysis with Sensory and Chemical Data

  • Keisuke Yoneda
  • Kimihiro Nakano
  • Keiichi Horio
  • Tetsuo FurukawaEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11306)

Abstract

In this paper, we propose a method for simultaneous analysis of subjective and objective data. The method, named coupled tensor self-organizing map (SOM), consists of two tensor SOMs, one of which learns the subjective data while the other learns the objective data. The coupled tensor SOM visualizes the dataset as three maps, namely, one target object map, and two survey item maps corresponding to the subjective and objective data. This method can be further extended to generate extra maps such as a map of attributes. In addition, the coupled tensor SOM also provides an interactive visualization of the relationship between the target objects and the survey items by coloring these three maps. We applied our proposed method to the wine aroma dataset. Our results indicate that this method facilitates an intuitive overview of the dataset.

Keywords

Subjective and objective data Multi-view data Coupled tensor decomposition Multi-relational data Tensor SOM Self-organizing map 

References

  1. 1.
    Acar, E., Nilsson, M., Saunders, M.: A flexible modeling framework for coupled matrix and tensor factorizations. In: European Signal Processing Conference, pp. 111–115, no. 1 (2014)Google Scholar
  2. 2.
    Cheng, Y.: Convergence and ordering of Kohonen’s batch map. Neural Comput. 9(8), 1667–1676 (1997)CrossRefGoogle Scholar
  3. 3.
    Damianou, A.C., Ek, C.H., Titsias, M.K., Lawrence, N.D.: Manifold relevance determination. In: Proceedings of the 29th International Conference on Machine Learning, ICML 2012, vol. 1, pp. 145–152 (2012)Google Scholar
  4. 4.
    Dang, J., et al.: Optimal design of on-center steering force characteristic based on correlations between subjective and objective evaluations. SASAE Int. J. Passeng. Cars Mech. Syst. 7(3) (2014)CrossRefGoogle Scholar
  5. 5.
    Ek, C.H., Jaeckel, P., Campbell, N., Lawrence, N.D., Melhuish, C.: Shared Gaussian process latent variable models for handling ambiguous facial expressions. In: AIP Conference Proceedings, vol. 1107, pp. 147–153 (2009).  https://doi.org/10.1063/1.3106464
  6. 6.
    Fitzpatrick, R., Davey, C., Buxton, M., Jones, D.: Evaluating patient-based outcome measures for use in clinical trials. Health Technol. Assess. 2(14), i-74 (1998)Google Scholar
  7. 7.
    Graepel, T., Burger, M., Obermayer, K.: Self-organizing maps: generalizations and new optimization techniques. Neurocomputing 21, 173–190 (1998)CrossRefGoogle Scholar
  8. 8.
    Heskes, T., Spanjers, J.J., Wiegerinck, W.: EM algorithms for self-organizing maps. In: IJCNN, vol. 6, pp. 9–14 (2000)Google Scholar
  9. 9.
    Hotelling, H.: The relations of the newer multivariate statistical methods to factor analysis. Br. J. Stat. Psychol. 10(2), 69–79 (1957).  https://doi.org/10.1111/j.2044-8317.1957.tb00179.xCrossRefGoogle Scholar
  10. 10.
    Hu, L., Cao, J., Xu, G., Wang, J., Gu, Z., Cao, L.: Cross-domain collaborative filtering via bilinear multilevel analysis. In: IJCAI International Joint Conference on Artificial Intelligence, pp. 2626–2632 (2013)Google Scholar
  11. 11.
    Hu, Y., Loizou, P.: Evaluation of objective quality measures for speech enhancement. IEEE Trans. Audio Speech Lang. Process. 16(1), 229–238 (2008)CrossRefGoogle Scholar
  12. 12.
    Iwasaki, T., Furukawa, T.: Tensor SOM and tensor GTM: nonlinear tensor analysis by topographic mappings. Neural Netw. 77, 107–125 (2016).  https://doi.org/10.1016/j.neunet.2016.01.013CrossRefGoogle Scholar
  13. 13.
    Kirihiro, N.: Subjective analyses and their contrast to organoleptic and chemical detection of aroma components. Ph.D. thesis, Kyushu Institute of Technology (2016). (in Japanese)Google Scholar
  14. 14.
    Klami, A., Fi, A.K., Fi, S.J.V., Kaski, S., Fi, S.K.: Bayesian canonical correlation analysis. J. Mach. Learn. Res. 14, 965–1003 (2013)MathSciNetGoogle Scholar
  15. 15.
    Kwan, W.O., Kowalski, B.: Correlation of objective chemical measurements and subjective sensory evaluations. Wines of vitis vinifera variety ‘pinot noir’ from France and the United States. Analytica Chimica Acta 122(2), 215–222 (1980)CrossRefGoogle Scholar
  16. 16.
    Oberg, T., Sandsjo, L., Kadefors, R.: Subjective and objective evaluation of shoulder muscle fatigue. Ergonomics 37(8), 1323–1333 (1994)CrossRefGoogle Scholar
  17. 17.
    Salzmann, M., Ek, C.H., Urtasun, R., Darrell, T.: Factorized orthogonal latent spaces. J. Mach. Learn. Res. Proc. Track 9, 701–708 (2010)Google Scholar
  18. 18.
    Saw, A., Main, L., Gastin, P.: Monitoring the athlete training response: subjective self-reported measures trump commonly used objective measures: a systematic review. Br. J. Sport. Med. 50(5), 281–291 (2016)CrossRefGoogle Scholar
  19. 19.
    Seshadrinathan, K., Soundararajan, R., Bovik, A., Cormack, L.: Study of subjective and objective quality assessment of video. IEEE Trans. Image Process. 19(6), 1427–1441 (2010)MathSciNetCrossRefGoogle Scholar
  20. 20.
    Sun, S.: A survey of multi-view machine learning. Neural Comput. Appl. 23(7), 2031–2038 (2013)CrossRefGoogle Scholar
  21. 21.
    Verbeek, J., Vlassis, N., Krose, B.: Self-organizing mixture models. Neurocomputing 63, 99–123 (2005)CrossRefGoogle Scholar
  22. 22.
    Wang, Z., Bovik, A., Sheikh, H., Simoncelli, E.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)CrossRefGoogle Scholar
  23. 23.
    Weeden, J., Sabini, J.: Subjective and objective measures of attractiveness and their relation to sexual behavior and sexual attitudes in university students. Arch. Sex. Behav. 36(1), 79–88 (2007)CrossRefGoogle Scholar
  24. 24.
    Xu, C., Tao, D., Xu, C.: A survey on multi-view learning, pp. 1–59 (2013). ArXiv http://arxiv.org/abs/1304.5634
  25. 25.
    Xu, Z., Jiang, H., Kong, X., Kang, J., Wang, W., Xia, F.: Cross-domain item recommendation based on user similarity. Comput. Sci. Inf. Syst. 13(2), 359–373 (2016).  https://doi.org/10.2298/CSIS150730007ZCrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Keisuke Yoneda
    • 1
  • Kimihiro Nakano
    • 2
  • Keiichi Horio
    • 1
  • Tetsuo Furukawa
    • 1
    Email author
  1. 1.Kyushu Institute of TechnologyKitakyushuJapan
  2. 2.Kuraray Co., Ltd.TokyoJapan

Personalised recommendations