Hi, This is our experience of implementing F-OWL system and applying it for OWL test case. -------------------------------- F-OWL 11-23-2003 -------------------------------- ---About F-OWL F-OWL is an ontology inference engine for the OWL. The ontology inference mechanism in F-OWL is implemented using Flora-2, an object-oriented knowledge base language and application development platform that translates a unified language of F-logic, HiLog, and Transaction Logic into the XSB deductive engine. F-OWL provides Java API, Command Line interface and GUI to fit different requirement. F-OWL includes: *A set of rules that reason over the data model of RDF-S & OWL *A set of rules that map XML DataTypes into XSB terms *A set of rules that perform ontology consistency checks *A set of rules that "glue" together the upper Java API calls to the lower layer Flora/XSB rules ---Tools used *XSB, Flora as reasoning engine *Jena ARP as RDF/OWL parser *Yajxb as interface between Java and XSB. ---OWL test case *Use of OWL Namespace: Check OWL Syntax *Entailment Tests: Load Premises file, Prove Conclusion file is TRUE *Non-Entailment Tests:Load Premises file, Prove Conclusion file is FALSE *OWL for OWL Tests: Prove conclusion is TRUE *Consistency Tests: Run consistency checker, Prove the file can be satisfied. *Inconsistency Tests: Prove the file cannot be satisfied via consistency checker. *Import Entailment/Level Tests: load imported ontology, check for import loop. F-OWL fails on: *Some large test files (e.g., food and wine) *OWL DL Files with Complex XSD datatype: for example: "invR-N-times-M-to-1" ---Some Numbers *The development of F-OWL was about 3 months, taking into account that we had been working on projects related to DAML+OIL language and XSB engine for two years. *The owl test case was about 1 month Regards UMBC F-OWL team http://fowl.sourceforge.net **************************************** Zou Youyong yzou1@cs.umbc.edu zouyyong@yahoo.com Address( Home): 4858 W.Braddock Rd.#201 Alexandria VA 22311 Phone (Home):703-627-9640 (LAB):410-455-2837 ****************************************Received on Sunday, 23 November 2003 11:18:38 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:43:29 GMT