Community-driven & Work-integrated Creation, Use and Evolution of Ontological Knowledge Structures
S. Braun. Karlsruhe Institute of Technology, (2012)
The thesis aims to support the collaborative development of ontological knowledge structures by communities of knowledge workers in order to facilitate the organization and sharing of information within their domain.
One big challenge for today’s organizations and knowledge workers is the focused discovery of new information that is likely to be interesting and useful in order to generate new knowledge. But it is also the organization thus information that had once be found and
identiﬁed as such can be rediscovered and shared. This is not only about the information itself but also about the people behind who hold the knowledge and e. g., may quickly provide assistance if there are questions. Knowledge about competencies and capabilities of its employees also is an essential need for an organization and its development. This encompasses activities like team stafﬁng or identifying training needs.
Research on ontology-based semantic (web) applications has shown that ontologies are well-suited for organizing and retrieving relevant resources being it people or documents because they connect information resources with machine processable background knowledge. However in practice, ontology-based applications still haven’t made their
breakthrough. This might be traced back to the high effort and complexity of ontology development. On the other hand, folksonomy-based systems recently have proven to be agile and user-driven approaches for the same application area. They enable their users to collect, manage and share information resources in an easy and lightweight way.
However, their lack of semantics also causes a number of problems plaguing tagging and hampering tag-based retrieval. To that end, this thesis explores how we can combine folksonomy- and ontology-based
approaches so that we keep their particular advantages and avoid their disadvantages thus supporting communities of knowledge workers in organizing and maintaining a shared information repository. This is investigated in the application of Social Semantic Bookmarking and Semantic People Tagging.
We present Ontology Maturing as a new perspective and conceptual model for the collaborative development of ontological knowledge structures. It supports (1) the development of a shared understanding, (2) the translation of Web 2.0 approaches to ontology engineering for more active participation, (3) the incremental formalization, (4) application-orientation & work-integration and (5) usable evolving models. To that end, we analyze the advantages and challenges of ontologies and ontology-based knowledge organization systems and make a comparison and consolidation of ontology spectra in literature as
well as of ontology development methodologies and tools. A conceptual design framework complements the ontology maturing model. It supports
software developers in deriving and realizing socio-technical systems that scaffold and guide ontology maturing in the application of Social Semantic Tagging for a given organizational setting. It considers technical as well as non-technical aspects. It organizes and provides methods and tools with that end users without modeling expertise can
collaboratively organize their information with ontologies and develop the latter one in a work-integrated way. To that end, we analyze the advantages and challenges of folksonomies and folksonomy-based systems and classify tagging motivations and categories
in literature. On this basis, we develop a general deﬁnition and model of social semantic tagging and its specializations of social semantic bookmarking and semantic people tagging.
The SOBOLEO framework presents a ﬂexible culture-system-ﬁt framework and reference implementation of the conceptual model and conceptual design framework. It provides a conﬁgurable and extensible architecture as well as reusable reference data models for social semantic bookmarking and semantic people tagging and competence ontology maturing. The review of related work shows that SOBOLEO is a pioneer for SKOS editors and the ﬁrst implementation for semantic people tagging ever.
Following the methodology of design-based research, the model, conceptual design framework and technical framework have been validated and iteratively improved in nine case studies with more than 250 participants involved.