Word Sense Disambiguation (WSD) is one of the topics where The Stanford Natural Language Processing Group is focused on. It’s relation with the translation issue.

In computational linguistics, word sense disambiguation (WSD) is the process of identifying which sense of a word (having a number of distinct senses) is used in a given sentence. For example, consider the word bass, two distinct senses of which are:

1. A type of fish

2. Tones of low frequency

The automatic disambiguation of word senses has been an interest and concern since the earliest days of computer treatment of language in the 1950’s. Sense disambiguation is an” intermediate task” (Wilks and Stevenson, 1996) which is not an end in itself, but rather is necessary at one level or another to accomplish most natural language processing tasks. It is obviously essential for language understanding applications such as message understanding, man-machine communication, etc.; it is at least helpful, and in some instances required, for applications whose aim is not language understanding:

·        Machine translation: sense disambiguation is essential for the proper translation of words.

·         Information retrieval and hypertext navigation: when searching for specific keywords, it is desirable to eliminate occurrences in documents where the word or words are used in an inappropriate sense.

·        Content and thematic analysis: a common approach to content and thematic analysis is to analyze the distribution of pre-defined categories of words across a text.

·        Grammatical analysis: sense disambiguation is useful for part of speech tagging.

·        Speech processing: sense disambiguation is required for correct phonetization of words in speech synthesis.

·        Text processing: sense disambiguation is necessary for spelling correction.

There’s are some WSD paradigms that have been proposed for machine translation (MT), which are:

·        Knowledge-based approaches: depend on manual linguistic knowledge and disambiguation rules.

·        Corpus-based approaches: make use of knowledge taken from text using machine learning techniques.

·        Hybrid approaches: mix characteristics of the two previous ones.

Nowadays, the most used ones in the recent works are the corpus-based and the hybrid techniques because they have very good results.

 

 

Sources:

·        Wikipedia, la enciclopedia libre. Última modificación: 03-05-08. Fecha de consulta: 07-05-08 from http://en.wikipedia.org/wiki/Word_sense_disambiguation

·        Ide, Nancy; Veronis, Jean: (1998), Word Sense Disambiguation: The State of the Art. Retrieved 17:28, May 13th, 2008 from http://sites.univ-provence.fr/veronis/pdf/1998wsd.pdf