omega-0.9.5
- From the Ontology Development Effort at the Intelligent Systems Division of the USC/Information Sciences Institute. This effort includes the original SENSUS ontology developed by Kevin Knight, Eduard Hovy, and Richard Whitney for machine translation and its Ontosaurus Web Browser, developed by Ramesh Patil and William Swartout. Our current effort is the OMEGA successor ontology by Eduard Hovy, Michael Fleischman, and Andrew Philpot, which supports information integration and ontology alignment as well as machine translation. OMEGA can be viewed using the DINOmega browser by Andrew Philpot. Using DINOmega to view the OMEGA Ontology DINOmega is the name of the browser with which you can explore OMEGA. OMEGA is a 120,000-node concept thesaurus (ontology) based on MIKROKOSMOS, the Penman Upper Model (built at ISI in the 1980s) and WordNet (built at Princeton University in the mid-1990's by George Miller and colleagues); Kevin Knight at ISI rearranged and subordinated WordNet to the Penman Upper Model at ISI. DINOmega is a reimpelementation of ONTOSAURUS, which was built at ISI by Ramesh Patil and Tom Russ. Each node is OMEGA represents one concept, i.e., one specific sense of a word. (Many words in English have many senses: "shoe" is the thing you wear on your foot, part of a brake, the action of nailing a bent piece of metal to a horse's hoof, etc.) The concepts are linked in a straightforward IS-A hierarchy, becoming ever more general as you go upward toward the root of the ontology. The top of the ontology is OB-THING, which is split into OBJECTs, PROCESSes, QUALITYs, and so on. OMEGA is not a strict tree--there are some multiple inheritance links. OMEGA is a so-called Terminological Ontology--although some links in OMEGA express part-of, synonym, antonym, and a few other relations between concepts, it does not at this time include much semantic information. (A more semanticized ontology would contain axioms that express all kinds of world knowledge, such as that cars typically have 4 wheels, that an airplane is a thing that flies and carries people, that animals do not talk, and so on.) Projects at ISI are underway to extract some such information from other sources of information such as encyclopedias or text. To explore DiNo, you need access to the Internet. Activate Netscape (or some other browser), and open http://blombos.isi.edu:8000/ You should see a two-pane display, with a large green dinosaur. The left pane displays the lexical thesaurus (this helps you go from English word, which may have many meaning senses, to a unique concept). The right pane displays the concept, its superclasses and subclasses and siblings. As usual, anything in blue and underlined is mousable. Enter your search term in the input window. If you click Word, DiNo will search for the word entry for the term you entered and display its senses in the left window. From this, you can select one or more concepts. If you type in a concept symbol instead of a word, please click on Concept. You may also type in a substring fragment and browse the concepts and words matching that fragment, by use of the Match button. DINO has two methods of showing to which ontology a concept belongs. In "prefixed" mode, a concept or word outside the "home ontology" will be prefixed with its ontology name, such as WORDNET::product. In "color" mode, each word or concept will be coded in a particular color; click on Legend to see the color mapping More information about OMEGA and its research predecessor SENSUS: Source of SENSUS--WordNet: - Miller, G.A. (1990). WordNet: An Online Lexical Database. International Journal of Lexicography 3(4) (special issue). Topmost Upper Structure of SENSUS: - Bateman, J.A., Kasper, R.T., Moore, J.D., and Whitney, R.A. 1989. A General Organization of Knowledge for Natural Language Processing: The Penman Upper Model. Unpublished research report, USC/Information Sciences Institute, Marina del Rey, CA. Creating and linking SENSUS to other ontologies and lexicons: - Knight, K. and S.K. Luk. 1994. Building a Large-Scale Knowledge Base for Machine Translation. Proceedings of the AAAI Conference. - Hovy, E.H. 1998. Combining and Standardizing Large-Scale, Practical Ontologies for Machine Translation and Other Uses. Proceedings of the International Conference on Language resources and Evaluation (LREC). Granada, Spain. Uses of SENSUS: - Swartout, W.R., P. Patil, K. Knight, and T. Russ. 1996. Toward Distributed Use of Large-Scale Ontologies. In Proceedings of the 10th Knowledge Acquisition for Knowledge-Based Systems Workshop. Banff, Canada.