The realization of the envisioned Semantic Web is conditioned by a real operationalized sharing and reuse of formally represented knowledge models, so-called ontologies. However, because of the open, distributed nature of Web environment and the inherent limitations of achieving highly reusable, commonly agreed domain conceptualizations, the current state of the art in the ontology engineering field states the need for additional techniques to aid ontology designers and users in evaluating and manipulating existing ontological sources for reuse purposes. The first goal of this work is to identify the major factors which influence the feasibility of current ontology reuse processes. We perform this risk analysis by means of a comprehensive literature survey complemented by two case studies situated in typical application scenarios in the Semantic Web field. The results of this analysis are aligned to specific contextual dimensions: features of the ontologies to be reused and of the corresponding ontology application scenarios, as well as parameters of the reuse engineering process. Accounting for the conclusions of the feasibility study, we advocate that the success of reuse endeavors is fundamentally influenced by factors situated at these dimensions and design an context-sensitive ontology reuse methodology implementing this finding. Our second goal is to provide methods and tools assisting ontology engineers and domain experts in operating reuse. On the basis of the requirements derived from the case studies and from the proposed methodology we develop a metadata model for the description of Semantic Web ontologies. This model represents reuse-relevant characteristics of ontologies in a semantically precise, machine-processable, interoperable and extendable manner. We elaborate on the context-sensitive usage of metadata information for human-driven ontology evaluation and exemplify it on a further sub-task of the reuse process, the matching between heterogeneous ontologies—as often encountered on the current Semantic Web. We also address the issue of metadata management: we propose a suite of heuristics to automatically derive and detect metadata information related to ontologies, thus minimizing the need for a manual annotation of the resources. The methods are prototypically implemented in a dedicated ontology reuse platform. In order to validate our research from different perspectives, we apply multiple evaluation approaches. The case study methodology provides us the theoretical foundations for comparing the operation of reuse processes in real-world situations with and without our context-oriented methodological approach, and for demonstrating its practicability. Professional reviews estimate the technical quality, originality and impact of the proposed solution. Using the goal-free evaluation methodology we are able to situate the results of this thesis in the actual ontology reuse field, emphasizing the scenarios in which they can be optimally applied. The implementation is evaluated against method-specific quality criteria on representative test sets.