Words can have more than one meaning, i.e., they can be polysemous. For example, the word glass can refer to a drinking glass and to the glass as a material. In a given context, the meaning of a word is usually clearly defined and thus unambiguous. Although it is obvious for humans that the word glass in the sentence “Sarah is drinking a glass of orange juice” is used in its meaning of a drinking glass, it is not easy for a computer to define the appropriate meaning to that word. The automatic identification which sense of a word (i.e. meaning) is used in a sentence is described as the problem of Word Sense Disambiguation (WSD). Mainly, there are two approaches for this task: WSD that is based on a sense inventory, a lexical resource such as a dictionary or a wordnet, which contains a predefined set of word senses, aims at assigning one sense from the sense inventory to the word to be disambiguated. Other WSD approaches are corpus-based, where the collocates of a word are used to define the words’ meaning.