Hypermedia presentation generation for semantic web information systems

Due to Web popularity many information systems have been made available through the Web, resulting in so-called Web Information Systems (WIS). Due to the complex requirements that WIS need to ful??ll, the design of these systems is not a trivial task. Design methodologies provide guidelines for the construction of WIS such that the complexity of this process becomes manageable. Based on the separation-of-concerns principle some of these methodologies propose models to specify di??erent aspects of WIS design. Model-driven WIS design methodologies have been recently in uenced by emerging technologies like the ones provided by the Semantic Web. We call WIS that use Semantic Web technologies Semantic Web Information Systems (SWIS). Hera is a SWIS design methodology that employs RDF, the foundation language of the Semantic Web, for its model representation. Using a standardized language to represent models fosters application interoperability. There are two main phases in Hera: the data collection phase, which makes available data coming from di??erent possibly heterogeneous sources, and the presentation generation phase, which builds Web hypermedia presentations based on the previously integrated data. This dissertation concentrates on the presentation generation phase of the Hera methodology. Chapter 1 introduces the research questions and provides an outline of the dissertation content. Chapter 2 gives an overview of existing model-driven (S)WIS design methodologies and their support tools. It also identi??es a number of desired (S)WIS design methodology features that are used as a comparison criteria for the analyzed methodologies. Chapter 3 describes the presentation generation phase of Hera, a model-driven SWIS design methodology. Di??erently than many of the analyzed SWIS design methodologies, Hera has most of the desired features of a SWIS design methodology like data integration support, presentation personalization, and user interaction. All Hera models and their instances are described in RDF. The presentation generation phase of the Hera methodology identi??es the following design steps: ?? conceptual design: constructs the conceptual model (CM), a uniform representation of the application's data. It de??nes the concepts and the concept relationships that are speci??c to the application's domain. ?? application design: constructs the application model (AM), the navigational structure over the application's data. It de??nes slices and slice relationships. A slice is a meaningful presentation unit of some data. There are two types of slice relationships: navigation relationships, used for links between slices, and aggregation relationships, used for embedding a slice into another slice. ?? presentation design: constructs the presentation model (PM), the look-and-feel speci ??cations of the presentation. It de??nes regions and region relationships. A region is an abstraction for a rectangular area on the user display where the contents of a slice are presented. As for slices, there are two types of region relationships: navigation relationships, used for links between regions, and aggregation relationships, used for embedding a region into another region. Regions have associated layout (positioning of inner regions inside a region) and style (fonts, colors, etc.) information. There are two types of presentation adaptation supported in the presentation generation phase of Hera: static adaptation, i.e., adaptation performed before the user starts browsing, and dynamic adaptation, i.e., adaptation performed while the user is browsing. The static adaptation is based on appearance conditions for elements in the CM, AM, and PM. These conditions use data from a user pro??le (UP) which stores the static user preferences and device capabilities. The dynamic adaptation uses AM queries in order to update the user session (US). US stores dynamic data, i.e., data created during user browsing based on user's input. The presentation generation process uses data from US in a similar way as the CM. Based on these two types of adaptation there are two variants of Hera's presentation generation phase: the static variant and the dynamic variant. Chapter 4 describes a CASE tool, the Hera Presentation Generator (HPG), that supports the presentation generation phase of Hera. There are two variants of the HPG, HPGXSLT, which corresponds to the static variant of Hera's presentation generation phase, and HPG-Java, which corresponds to the dynamic variant of Hera's presentation generation phase. HPG-XSLT implements the data transformations using XSLT stylesheets, and HPG-Java implements the data transformations in Java using Jena and Sesame libraries. HPG-Java exploits more of the RDF model semantics than HPG-XSLT. A distributed architecture of the HPG based on Web Services is also provided. Chapter 5 proposes RAL, an RDF algebra that can be used for RDF query optimization. It de??nes a data model and a set of operators. The collections (sets) of nodes are closed under all operators. There are two types of operators: extraction operators, which retrieve nodes of interest from an input collection, and construction operators that build an output model possibly using also the extracted nodes. The extraction operators satisfy equivalence laws resembling to the ones found in relational algebra. We propose a heuristic algorithm for RDF query optimization similar to the one given in relational algebra. Chapter 6 shows how one can apply a general-purpose graph visualization engine, GViz, for the visualization of the models used in the presentation generation phase of Hera. Compared with other visualization tools GViz has the advantage of being easily customizable. As Hera models have the tendency to be rather large, we de??ne visualization scenarios in order to get a better insight into the model properties. We did use GViz for the visualization of conceptual models, conceptual model instances, application models, and application model instances.