Pasin, M., Motta, E., & Zdrahal, Z. (2007). Capturing knowledge about philosophy. In Proceedings of the 4th international conference on Knowledge capture (pp. 47-54). Whistler, BC, Canada: ACM.
Goals of Research:
- Development of an ontology to support annotating and reasoning about philosophical knowledge
The study presents an ontology developed for the domain of philosophy. The rationale of this development is to use this ontology for annotating and reasoning about philosophical knowledge and providing dynamic organization of learning material with a high level of flexibility. The study addressed the modeling problems and considerations for this specific domain of knowledge inline with issues with natural language ambiguities. PhiloSURFical is a tool developed during this study to support ontology-driven exploration of of philosophical resources. Through discussions with domain experts and analysis of curricula of philosophical textbooks, three generic features of philosophical domain has been recognized: 1) it is a vast domain (historical and theoretical dimensions); 2) it is dynamic and slippery; 3) it is based on argumentation practice (exegesis of a text, interpretation of an idea, and discussion about meaning) (p. 48). Therefore, the subtile semantic model for this domain should provide support for representation of: historical events, generic uncertainty, information objects, interpretation events, contradictory information, viewpoints, and varying granularity (p.48). CIDOC-CRM is adopted as the base development ontology, however, to improve the coverage of the ontology it has been integrated with three other ontologies, AKT reference ontology, DOLCE ontology, and FRBR. The existing concepts and relations also extended to cover description of philosophical events and ideas (p. 49). The most important aspects in this study are the domain specific characteristics such as, Philosophical Events (intellectual movements, and social and intellectual activities) (pp. 49-51); Philosophical Documents (propositional-content, expression, representation-form, manifestation) (pp. 51-52); Philosophical Ideas (argument-entity, method, problem-area, problem, view, distinction, rhetorical-figure) (pp. 52-53). Finally an ontology-driven tool has been developed for navigation of philosophical resources, and the tool would be used for the task-based evaluation of ontology (p. 54).
- Analysis of domain and testing with a pilot study (Wittgenstein Tractatus).
- CIDOC-CRM is employed as the base ontology because it is adopted as an standard and also it is an event-base ontology (p.48)
- The ontology is developed by OCML and exported to OWL and Ontolingua (p. 49)
- CIDOC-CRM does not provide facilities for reasoning on time (p. 49). It is weak regarding modeling of social and intellectual activities (p. 50), and it is more suited for representing physical features of museum artifacts (p. 51)
- Clarification of the use of terms is very important in a domain modeling (p. 50)
- The use of “semantic relevance” of ideas (based on word co-occurrence) for modeling
- Task-based evaluation of ontology
- Niepert, M., C. Buckner, and C. Allen. A dynamic ontology for a dynamic reference work. in JDCL-2007. Vancouver, British Columbia, Canada.
- Welty, C. and J. Jenkins, An Ontology for Subject. Journal of Data and Knowledge Engineering, 1999. 31(2): p. 155-181.