Word Choice Problem and Anaphora Resolution

摘要

A dictionary-based method of implicit links between words in the texts(so-called indirect anaphora) is discussed. We suggest that the same dictionary and method can be used to partially solve the word choice porblem in machine translation.The method consists in using of a dictionary of "scenarios" -lists of words semantically related to the given one, and show that detecting the implicit referential relationships can be viewed as intersection of such scenarios. The advantage of the method is in the simplicity of the dictionary being used, since it does not rely on specific semantic relationships between the headword and the words listed in its scenario. Thus, such a dictinary can be derived from some existing semantic dictionaries or even from large corpora.