An object-oriented version model for context-aware data management

Cite This

Files in this item

Checksum: MD5:9a3708f64fa5cdfab093f5b4615f4976

GROSSNIKLAUS, Michael, 2007. An object-oriented version model for context-aware data management. Zürich : ETH Zürich

@book{Grossniklaus2007objec-25912, publisher={Zürich : ETH Zürich}, title={An object-oriented version model for context-aware data management}, year={2007}, author={Grossniklaus, Michael}, note={Zugl.: Diss., Univ., Zürich, 2007} }

An object-oriented version model for context-aware data management 2007 Grossniklaus, Michael 2014-01-24T10:12:33Z 2014-01-24T10:12:33Z Grossniklaus, Michael eng terms-of-use Context-aware computing has been recognised as a powerful solution to address the requirements of many modern application domains. For example, in mobile computing, context is used to augment the user's experience when interacting with a system by adapting responses to their current situation. In the domain of ubiquitous and pervasive computing, the role of context is often to compensate for the lack of traditional user interfaces. Finally, in web engineering, context is used to adapt content delivery to numerous factors. Building on context information, recent web systems support multi-channel and multi-modal interaction as well as personalisation and internationalisation. While several frameworks and models have been proposed to gather, represent and process context, very few of these solutions consider the management of context-aware data.<br />In the past, database systems have been used to manage data for applications from various domains. To address ever changing requirements,database systems have undergone a remarkable evolution from monolithic heavyweight databases to lightweight personal information stores. Continuing this evolution, it is our approach to extend database systems with concepts to cope with the challenge of context-aware data management. The proposed solution is based on a two-dimensional version model that allows context-dependent variants to be managed, while, at the same time, keeping track of the revisional history. Within the scope of this version model, query processing is realised based on a matching algorithm that uses the current context state of the system to select the best version of every object that is accessed during query evaluation. Our version model has been specified based on the OM object-oriented data model and has been implemented as part of a database management system that is based on OM. As an application of this extended data base management system, a web content management system has been designed and implemented. The intention of this Extensible Content Management System (XCM) is to provide a powerful implementation platform for web engineering applications. Most model-driven design methodologies in this application domain feature some notion of contextdriven adaptation. However, suitable implementation platforms to support these methodologies are sparse or tied in with a specific design method. To implement web engineering systems, XCM separates the concepts of content, structure, view and presentation. Metadata about these concepts is managed within our extended database and therefore all four aspects of a web system can be made context-aware.<br />The EdFest mobile tourist information system serves as a proof-ofconcept for XCM, that is used as a content management component in this system. EdFest is an application that provides information about the Edinburgh Festivals to tourists while they are visiting the city. The system offers multi-channel interaction by supporting standard web channels as well as a novel paper-based channel. In terms of context information, EdFest has established new requirements, as not only traditional factors such as location, device and user context are considered, but also the interaction context. Due to specific requirements of the interactive paper delivery channel, not only the content that is delivered to a tourist has to be adapted but also the interaction process that leads to the content delivery.<br />Even though our version model for context-aware data management has been defined within the framework of the OM data model, the concepts can be generalised to data that is represented in other models. In particular this is the case for relational database management systems, as our version model has its origins in solutions that have been proposed for temporal and engineering databases that were developed based on the relational model. Unfortunately, most of today's information is not managed by database systems but rather stored in plain files and it is<br />therefore not possible to take advantage of the proposed version model. Concepts from web content management such as the explicit representation of the internal structure of a file or the reuse of different content components in different places could be used to improve today's file systems. Only in this setting, can the full potential of context-aware data management be finally unleashed in handling everyday information such as wordprocessing, spreadsheet or presentation documents. Zürich : ETH Zürich

Downloads since Oct 1, 2014 (Information about access statistics)

Grossniklaus_259129.pdf 462

This item appears in the following Collection(s)

Search KOPS


My Account