Advanced search
Start date

Building a parallel corpus annotated with co-reference for machine translation


In spoken or written natural language, it is often the case that we make use of referring expressions to denote the real world objects, people, things, events, etc., that we talk about. In the form of proper nouns (e.g., John), pronouns (e.g., him), definite descriptions (e.g., the blue building on the left), indefinite descriptions (e.g., a man dressed like the president) and many others, referring expressions convey an important part of natural language meaning and, accordingly, have received a great amount of attention in the NLP field. The present proposal focuses on the development of a basic language resource: a parallel corpus annotated with correference information for the computational treatement of referring expressions. The corpus should address the needs of a particular kind of application, namely, machine translation systems for Portuguese language, whilst being sufficiently generic to support the future development of other NLP applications. In this sense, we believe that the proposed work is an ideal first NLP project in this University. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
Articles published in other media outlets (0 total):
More itemsLess items

Please report errors in scientific publications list by writing to: