Service Mining by Wei-Lun Chang

Author:Wei-Lun Chang
Language: eng
Format: epub
ISBN: 978-1-60649-575-9
Publisher: Business Expert Press
Published: 2014-01-16T16:00:00+00:00

Typical student

Addicted student

Typical worker

Addicted worker

B

C

B

C

E

B

C

E

S

B

C

E

S

26

0.734

0.833

0.723

0.667

0.753

0.792

1

0.667

1

0.225

0.778

0.89

39

0.816

0.793

0.734

0.835

0.835

0.778

0.793

0.667

1

1

0.558

0.736

0.906

52

0.793

0.89

0.778

0.89

0.835

0.76

0.834

0.667

1

1

0.505

0.667

0.752

65

0.787

0.81

0.730

0.89

0.802

0.784

0.862

0.89

1

1

0.604

0.81

0.918

In step 2, the results revealed that 52 days was optimal for typical students and addicted students. In addition, the optimal period of study for typical workers and addicted workers was 65 days. This means that enterprises would have to collect more data to predict the behavior of typical and addicted workers. In particular, the performance of the Bayesian probability model increased with an increase in the length of the study period. In summary, this research used a two-step prediction for customer value and CRM e-service combinations. Simulation results revealed that the number of days should be large enough to improve performance. That is, the data should be no less than 52 days (i.e., 2 months) for either the Markov chain or the Bayesian theorem. Segmentation of customers could also help firms to identify customer characteristics and discover how to deal with their needs.

In this chapter, the concept of service mining provides a new viewpoint to reexamine CRM on electronic services. Customer behavior is difficult to predict especially on Internet. Thus, service mining can help firms to simplify customer behavior on services and predict potential services in advance. Human beings always repeat 80% of behavior according to psychology. That is, enterprises may identify key successful and most used services by modeling the behavior on services. The service will be refined and reorganized to earn more benefit and profit.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.