Evaluating the performance of the OntoSem semantic analyzer
TextMean '04 Proceedings of the 2nd Workshop on Text Meaning and Interpretation
Semantically rich human-aided machine annotation
CorpusAnno '05 Proceedings of the Workshop on Frontiers in Corpus Annotations II: Pie in the Sky
IWIC'07 Proceedings of the 1st international conference on Intercultural collaboration
Hi-index | 0.00 |
In this paper we compare programs of work that aim to develop broad coverage cross-linguistic resources for NLP: Ontological Semantics (OntoSem) and SIMPLE. The approaches taken in these projects differ in three notable respects: the use of an ontology versus a word net as the semantic substrate; the development of knowledge resources inside of as opposed to outside of a processing environment; and the development of lexicons for multiple languages based on a single core lexicon or without such a core (i.e., in parallel fashion). In large part, these differences derive from project-driven, real-world requirements and available resources -- a reflection of their being practical rather than theoretical projects. However, that being said, we will suggest certain preferences regarding the content and development of NLP resources with a view toward both short- and long-term, high-level language processing goals.