This paper describes MECUREO, a system that automatically constructs an ontology from a dictionary. We describe its use in the context of a dictionary of computer science. MECUREO parses this to construct a graph representing a lightweight ontology of computer science. This paper describes MECUREO’s approach to parsing the dictionary and the construction of the ontology graph. We also describe MECUREO’s point query facility. This generates a small ontology for use in a particular teaching context, such as single lesson. We report two evaluations of the effectiveness of the ontology construction, one based on a qualitative concept mapping experiment and the other a quantitative comparison of the computer science ontology with another similar ontology.

ION6. 3 The IWN Project The basic premises of IWN are as follows: ffl A concept ontology designed for American English can also be used as the basis for organising word senses in other languages; ffl Each word sense can be defined with respect to a set of contexts in a corpus rather than by means of the traditional definition or `gloss'; We are investigating three paradigms by which source word senses can become associated with WordNet word senses: Paradigm A: A lexicographer is provided with a word sense in the source language, e.g. RU.OFIS and they must provide a WordNet synset to which it is to be mapped, e.g. EN.OFFICE -- (where professional or clerical duties are performed) Paradigm B: A lexicographer is provided with a synset in WordNet, e.g. EN.OBJECT, EN.INANIMATE OBJECT, EN.PHYSICAL OBJECT, -- (a nonliving entity) and they must provide a suitable word sense in the source language to which it corresponds, e.g. file, data file -- (a set of related records kept together) =...

this article we report on a study in which Russian word senses are mapped on to nodes of the WordNet ontology. In the following section we briefly describe WordNet. Following this a short explanation of the IWN project is given. Next we describe the Russian word sense mapping study which has been carried out. Finally we provide some conclusions and pointers for further work.