Modelling and analysis of web-based information systems

As the use of the World Wide Web (WWW) becomes popular, a lot of information is now distributed by means of HTML documents, a form of hypertexts. However, finding desirable information on the Web can be time consuming and ineffective. The complexity of the Web makes it dificult for users to get a full picture of the hyperlinks in which information is organised. The problem is aggravated with the incorporation of dynamic behaviour into information structure using facilities supported by JavaBeans and CORBA. This motivates the need to model HTML documents, as an information system, to facilitate rigorous analysis. Previous modelling in this area concentrated on the aspects of structures and contents of hypertexts only, whereas dynamic components (like Javascript and CGI), which are peculiar to HTML documents and are interrelated with contents and structures, received little attention. In this report, we propose a framework in CCS to unify the modelling of HTML documents based on structure, content and dynamic behaviour. Properties are then expressed as modal u-calculus formulae and checked against the model. Interesting queries regarding the document, like accessibility, ordering, reachability and mirror site verification, can be answered using automated tools such as the Concurrency WorkBench