Even though the use of electronic ontological knowledge representations dates back until the 90s, their usage today is still unabated with application scenarios not only in the sharing of information but also in the inferring of new knowledge through reasoning or the enablement of natural language processing. But as ontological representations are on the rise – how can one ensure the quality and correctness of such artifacts?
The usage of objective, quantifiable metrics provides reliable ontological measurements. And in the past, an enormous amount of metrics has been proposed, assessing – among others - the graph and schema attributes, annotations, relations, or instances. But often, their influence on the actual quality is not researched extensively. How, and in what kind of composition do the metrics proposed by Tartir et. al. in the OntoQA-Framework influence the understandability of an Ontology? How does Gangemi et. al.’s graph metrics influence the reusability of an ontology? The influence of certain metrics on concrete quality attributes is often not described and if so, not validated in an empirically sound approach. Further, most of the metrics stay rather isolated. It is often not known how metrics correlate with each other. These shortcomings make the usage of ontology metrics arbitrary – especially inexperienced modelers are facing challenges selecting the right metrics for the right goals. Even though the ontology metrics are calculated objectively, the interpretation remains subjective.
Validated ontology quality measurements can help these modelers to develop ontological models based on their aimed usage scenario. A translation of the abstract measurements into high-level quality dimensions like, among others, “completeness”, “clarity” or “adaptability” helps to classify the own work into a broader context. This is especially true if these metrics are provided in a repository, allowing the comparison between the own creation and various other ontologies. Further, based on the provided quality calculations, possible improvements can be given for a certain quality goal, highlighting the artifacts that are the most influential factors for each quality dimension. In effect, this can not only lead to better ontologies but in the long term also to better-trained modeling staff.

The goal of this doctorate is to establish and validate a link between comprehensive quality measurements like “understandability” or “completeness” and the quality metrics proposed in the literature. Using a data-centric research design, the goal is the identification of quality grades and improvement recommendations. This has the potential to support especially inexperienced ontology engineers in assessing their work and the creation of better ontologies. The novelty of this research lays in the data-centricity of its design. Using a collection of large amounts of evolutional ontology metric data, statistical relevant correlations are to be found. This enables the validation of already proposed quality measurements and the identification of new ones.

As information technology increasingly permeates everyday life, it provides great potential and at the same time leads to new challenges. In the current working world two challenges can be observed in particular, a work intensification and blurring boundaries between personal and work life. Many employees are often pressed for time. Frequent interruptions and multitasking constitute additional difficulties for the planning and completion of tasks, both in business and personal life. The resulting stress combined with a lack of opportunities for recreation are hazards to cause ill health. Therefore, self-management is becoming increasingly important, not only with regard to productivity, but especially in terms of maintaining individual motivation, wellbeing, and health. Developing an individually fitting and frictionless workflow may entail additional effort, but can be a great benefit, especially in the long-term. Existing tools, such as digital calendars or to-do lists, offer only rudimentary support as most of them are static and require continuous manual adjustment. Moreover, developments in the area of sensor technology and smart devices are rarely taken into consideration so far. These technologies enable collecting data on users and their environment. The information obtained, e.g. the location of a user, his motion, or biological parameters, as well as conclusions, e.g. about the current situation, can be integrated into approaches to support self-management.

The aim of this research project is to develop and test a concept of an innovative assistance system for context-sensitive self-management. In particular, the potential of sensors and smart devices shall be examined to collect necessary data and to implement a ubiquitously usable recommendation feature. The results shall be used to develop a personalised, situation-aware, and stress-sensitive assistance system for self-management.