Contact us
[email protected] | |
3275638434 | |
Paper Publishing WeChat |
Useful Links
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Article
Purchase Analysis Based on the Relationship Between Customers and Service Providers
Author(s)
Yuya Miyamoto, Michiko Tsubaki
Full-Text PDF XML 1380 Views
DOI:10.17265/2328-2185/2018.02.001
Affiliation(s)
The University of Electro-Communications, Tokyo, Japan
ABSTRACT
Although the service industry takes up a large percentage in world
economies and is situated in the center of all industries, it has been pointed
out that its growth rate is smaller compared to the other industries, which is
partly because of its characteristic features: “amorphousness”, “simultaneity”, and “heterogeneity”. Service providers are
therefore required to be experienced and have a good business sense to succeed
in this field. The aim of this research is to support those who are not
experienced or do not have a good business sense, using scientific approach.
This research tries to present recommendation, taking customers’ value and
compatibilities with employees into account, and is based on the assumption
that customers and employees have one-to-one contact over a period. A total of 3,447 customers and 133 employees were classified according to their
philosophies, needs, and abilities. For
each case, customers’ purchase histories are first interrelated with the result of questionnaire, and
put purchase behavior into marketing using Bayesian network. Then the HUB was
examined and extracted features of the data of the questionnaire, and executed
the stochastic inference to present the recommendation. This procedure enabled
us to extract the features of customers’ purchase behavior, and it is turned
out that compatibilities of customers and employees are more important than the
difference of their values and abilities.
KEYWORDS
heterogeneous, customer values, recommendation, service science, purchase Bayesian network, stochastic reasoning
Cite this paper
References