An experimental study for some supervised lexical disambiguation methods of arabic language

In this paper we propose an experimental study for some supervised algorithms to disambiguate arabic words. Due to the lack of linguistic data for the Arabic language, we work on non-annotated corpus and with the help of four annotators; we were able to annotate the different samples containing the ambiguous words. Since that, we test the naïve Bayes algorithm, the decision lists and the exemplar based algorithm. During the experimental study, we test the influence of the window size on the disambiguation quality, the derivation and the technique of smoothing for the (2n+1)-grams. We find that the exemplar based algorithm achieves the best rate of precision.