Transformation of Web Event Sequences for Analysis of Users' Web Operation

This paper describes a method for transforming a user's Web operation sequence. The method handles a user's Web operation recorded as a sequence of events, and transforms the recorded event sequence into an event sequence suited for later analysis. The types of events include mouse movement, URL transition, button selection, text input to a form, window scrolling, and window sizing. The recorded data are represented as a sequence of events, and contents are separately stored in a Web server in a normal way. The recorded event sequence can be played with a Web browser. There are two major ways for analyzing the recorded operation. One way is analysis by a human who looks at the automatic playback. The other way is analysis by software. In both ways, transformation of an event sequence is required for effective analysis. The method in this paper provides three kinds of transformation; (i) changing a sequence of Web pages so that a transformed page sequence conforms to a standard page sequence, (ii) scaling time spent in a Web page so that the transformed time becomes a standard time for the page, (iii) inserting a page with no operation, into the recorded sequence when the page was not accessed in a user's operation and the page is included in a standard sequence.

By: Amane Nakajima and Yoshinori Aoki

Published in: 1999 IEEE International Conference on Systems, Man, and Cybernetics, Japan, IEEE, vol.4, p.111-116 in 1999

Please obtain a copy of this paper from your local library. IBM cannot distribute this paper externally.

Questions about this service can be mailed to reports@us.ibm.com .