The objective of this study is to develop the usage pattern model of web-based electronic commerce system under the consideration of end-user's usability for their convenience.
The procedures of this study are summarized as follows :
(1) Selection of the 10 shopping mall in Korea.
(2) Definition of the activity and procedure of decision making through the consideration of design form, link type, link method, type of screen design, and type of structure design in the shopping mall.
(3) Selection of the 34 variables, and analysing current C shopping mall.
(4) Making the log file 48(20~40 years) subjects participated in the experiment, and they randomly select one for 4 different kinds of goods.
(5) Data mining
(6) Principle component and cluster analysis (using the statistical package, SAS 8.1)
For whole user's usage pattern, the purchasing operation process is summarized as eight activities(status) by Start of Session, Reference, Display, Help, Browser, Order, Mail and Others. These usage patterns can be classify into two groups that one is applicable to the activity for the order and the other is for help, browser and mail. The former showed activity of directive cashing process, the latter showed information acquisition process on the usage. So, it is necessary to distinguish the structure of menu and screen on purchasing process and information acquisition process.
This thesis investigated the usage pattern of the user. And Markov Chain was used to quantify usage pattern on the shopping mall. So it is helpful to construct custom-made shopping mall to consider user's requirement. These results are applicable to the design for the usage promotion of the web navigation system, the books information searching system and the diverse navigation engine system as well as the shopping mall.