Yet another INduction Yields to ANother Generalization

Welcome. This is the homepage for YINYANG system. This system contains the implementation of the methodology for Supervised Concept Learning in Description Logics. It is the main contribution of Luigi Iannone's Ph.D. dissertation (available upon request). We believe that Machine Learning can be very useful in order to overcome the bottleneck of manual ontology creation. Machine Learning has the following advantages w.r.t. manual ontology building:

• It is a formal process
• It is based on evidence coming from data rather than on personal skills and expertise
• Expert can come into play after the learning to refine its results (semi-automatic approach)

Below you can find:

• YINYANG Release Notes you should read before using YINYANG
• YINYANG Java implementation binary code
• Some examples of learning problems:
  • Basic Family

Please Notice This system is still in its early phase of development. It is not meant for any other use than research (both academic and not). The main purpose for making it available is to make it use by unbiased people that can spot inconsistencies and bugs that its developers could not. Therefore, if you think you found a bug, an inconsistency or simply something weird do not hesitate in contacting the person in charge for YINYANG (i.e.: Luigi Iannone). Any feedback is welcome (as far as it is polite :-)) but for more technically sound replies it is forseaable that you paste in your messages both the particular learning problem and the console log. YINYANG works with DIG 1.1 Description Logic Reasoning services and has been successfully tested with RacerPro (Commercial) and Pellet (Free for research use)

For more theoretical aspects of our approach for Supervised Concept Learning we suggest the following papers (more up to date bibliography will appear soon):

• Floriana Esposito, Nicola Fanizzi, Luigi Iannone, Ignazio Palmisano, and Giovanni Semeraro. Knowledge-intensive induction of terminologies from metadata. In Frank van Harmelen Sheila A. McIlraith, Dimitris Plexousakis, editor, The Semantic Web ISWC 2004: Third International Semantic Web Conference, Hiroshima, Japan, November 7-11, 2004. Proceedings, volume 3298 of LNCS, pages 411Ð426. Springer-Verlag Heidelberg.
• Luigi Iannone, Ignazio Palmisano: An Algorithm Based on Counterfactuals for Concept Learning in the Semantic Web. In Moonis Ali and Floriana Esposito (Eds.) Innovations in Applied Artificial Intelligence, 18th International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems, IEA/AIE 2005, Bari, Italy, June 22-24, 2005, Proceedings: pp.370-379