In this paper, we describe a user-oriented reliability model that addresses extensive web content, including many links, and that simultaneously addresses user dynamic navigation behaviour. Computation time is a problem in web systems because content can be tightly coupled. An alternative approach is to skip the consideration of the actual dynamic relationships but this is undesirable because the quality of a web system is bound to the success of a series of user navigations. These issues motivate our study from a user perspective. By parsing the logs, past navigations can be restored to form a flow graph. Our model adopts the idea of cyclomatic complexity to convert the graph to a maximum number of independent paths so that reliability can be computed by discounting the probabilities of unreliable paths. The application of this model can prioritise the fault removal sequence to assist in resource and effort allocation to improve reliability.
|Original language||English (US)|
|Number of pages||19|
|Journal||International Journal of Web Engineering and Technology|
|Publication status||Published - 2007|
All Science Journal Classification (ASJC) codes
- Computer Networks and Communications