The InPhO is a dynamic ontology constructed in a 3-step iterative process, described fully in our JCDL 2007 and FLAIRS 2008 papers.
- A small, hand-built formal ontology was created using the subject-area structure of the Stanford Encyclopedia of Philosophy (SEP) and other resources to identify major topics and sub-topics in the domain of philosophical ideas.
- Statistical methods are run over the entries in the SEP to identify likely relationships among terms.
- Feedback from experts knowledgeable in philosophy is used to assess the results of step 2. This feedback is then stored as statements in first-order predicate logic, and then passed to a logic program we have designed to classify ideas in optimal locations for reachability to all other ideas deemed highly-related by experts. The finished product of this inference task is then viewable on the InPhO Taxonomy.
For more information, check the InPhO FAQ.
