Noun compounds are a major challenge for the automatic analysis of English written text. A noun compound is a sequence of nouns acting as a single noun, such as "colon cancer tumor suppressor protein" or "carbon steel soup pot cover". Often at least partially lexicalised, such constructions encode implicit relations which tend to be hard for language processing software to understand. For example, olive oil *is extracted from* olives, while malaria mosquito *spreads* malaria. Noun compounds are abundant in written English. They comprise 3.9% of the tokens in the Reuters corpus and 2.6% in the British National Corpus, so they cannot be conveniently ignored. They also are highly productive: over half of the two-noun compound types in the BNC occur only once. Moreover, noun compounds cannot be enumerated in any static resource: it has been shown that static English dictionaries cover only 27% of the noun compounds that occur 10+ times in the BNC.
It is not surprising that noun compounds have attracted a lot of research interest in theoretical linguistics and in computational linguistics. There has been considerable progress in the theory and practice of their semantic interpretation in the last several years, as well as new insights into the process of compounding and its use in text processing applications. For example, a question-answering system might need to determine whether "protein acting as a tumor suppressor" is a good paraphrase for "tumor suppressor protein". An information extraction system might need to decide whether "neck vein thrombosis" and "neck thrombosis" could co-refer when used in the same document. A machine translation system facing the unknown noun compound "WTO Geneva headquarters" might benefit from being able to paraphrase it as "Geneva headquarters of the WTO" or "WTO headquarters located in Geneva". Given a query like "migraine treatment", an information retrieval system could use paraphrasing verbs like "relieve" and "prevent" for page ranking and query refinement.
We invite contributions on topics related to computational approaches to the semantics of noun compounds, including but not limited to the following areas:
- designing models, resources and tools for the syntactic and semantic interpretation of noun compounds;
- comparing and mapping between different semantic representations;
- evaluating the quality of noun compound interpretation systems;
- paraphrasing noun compounds;
- adapting linguistic theories to the computational interpretation of noun compounds;
- applying noun compound interpretation to various natural language processing tasks.
We seek original unpublished papers conforming to the JNLE format -- see the Submission page for details.