Verb Sense Annotation in News Texts in the CSTNews Corpus

One of the hardest problems in Natural Language Processing (NLP) is the lexical ambiguity, as words may express different senses depending on the context in which they occur. In NLP, Word Sense Disambiguation (WSD) is the task that aims at determining the proper meaning of a word in its context. In this task, the use of a sense annotated corpus is useful because this computational linguistic resource enables further study of the ambiguity phenomenon and the development and evaluation of WSD methods. This paper describes the verb sense annotation process in news texts in the CSTNews corpus, using Princeton WordNet as sense repository. Besides detailing the annotation process and its results, the contributions of this work include the availability of a linguistic resource that may be the basis for future research in WSD for Portuguese.