Abstract
Coding is a task that classifies an object to a corresponding code (or class). This is often required for survey data processing in the field of official statistics. Since the governmental survey has large number of objects and codes (or classes), and the release time of the survey result has to be strictly observed, the autocoding system is a key solution for improving data processing. For this autocoding system, mainly two types of methodologies have been developed. One is the use of the supervised classification methods including machine learning techniques and the other is rule-based methods. For the supervised classification method, we have developed a supervised multiclass classifier using machine learning which has the advantages of simplicity and practical calculation time. In this paper, we present an application of the proposed method for the Family Income and Expenditure Survey in Japan with a comparison of the accuracy and the efficiency of the rule-based method.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Gweon, H., Schonlau, M., Kaczmirek, L., Blohm, M., Steiner, S.: Three methods for occupation coding based on statistical learning. J. Off. Stat. 33(1), 101–122 (2017). https://doi.org/10.1515/jos-2017-0006
Hacking, W., Willenborg, L.: Method series theme: coding; interpreting short descriptions using a classification. In: Statistics Methods. Statistics Netherlands (2012). Available at: https://www.cbs.nl/en-gb/our-services/methods/statistical-methods/throughput/throughput/coding. Cited 16 Nov 2017
Kudo, T., Yamamoto, K., Matsumoto, Y.: Applying conditional random fields to Japanese morphological analysis. In: Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing (EMNLP-2004), pp. 230–237 (2004)
Shimono, T., Wada, K., Toko, Y.: A supervised multiclass classifier using simple machine learning algorithm for autocoding. Res. Mem. Off. Stat. 75, 41–60 (2018) (in Japanese)
Taguchi, G.: Mathematical for quality engineering – 7. Signal-to-noise ratio for chemical and biological systems. In: Quality Engineering Forum 5(2), 3–9 (1997) (in Japanese)
Toko, Y., Wada, K., Kawano, M.: A supervised multiclass classifier for an autocoding system. J. Rom. Stat. Rev. 4, 29–39 (2017)
Yui, S.: Application of a statistical learning algorithm to autocoding system. Statistics, Jan. 2017. Japan Statistical Association (2017) (in Japanese)
Acknowledgements
We are grateful to Dr. Tusbaki, H., Director-General of the Institute of Statistical Mathematics for helpful comments for this research.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Toko, Y., Wada, K., Yui, S., Sato-Ilic, M. (2020). A Supervised Multiclass Classifier as an Autocoding System for the Family Income and Expenditure Survey. In: Imaizumi, T., Okada, A., Miyamoto, S., Sakaori, F., Yamamoto, Y., Vichi, M. (eds) Advanced Studies in Classification and Data Science. Studies in Classification, Data Analysis, and Knowledge Organization. Springer, Singapore. https://doi.org/10.1007/978-981-15-3311-2_40
Download citation
DOI: https://doi.org/10.1007/978-981-15-3311-2_40
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-3310-5
Online ISBN: 978-981-15-3311-2
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)