Usability Analysis

Research in human-computer interactions (HCI) has recently started developing a computational science of usability [GSB94]. The principal goal of this effort is develop a systematic approach to usability studies by adapting the rigorous experimental method of a computational science. The first step is to develop instrumentation methods which collect data about software usability, in a manner akin to instrumentation that has been done for analyzing performance. This data is then used to build computerized models and simulations which explain the data. Finally, various data presentation and visualization techniques are used to help analyst understand the phenomenon. This approach can also be used to model the browsing behavior of users on the Web.

As described in this section, there is an increasing need for, as well as interest in, developing techniques and tools to analyze the usage patterns of information on the Web. Some initial ideas have been proposed, but are still in their nascent stages and much work remains to be done. We believe that the techniques which are most effective will include the following characteristics: (i) will be data driven empirical methods, (ii) will use vast amounts of data for validation, (iii) will use rigorous experimental methods and sound statistical analysis, etc.