Word sense ambiguity is a potential source of error in many areas of computerized language analysis, like information retrieval, knowledge acquisition and machine translation. Text classification, as a subfield of information retrieval, is also believed to suffer from this phenomena, and therefore assumed to benefit from word sense disambiguation (WSD).

This talk reports on a number of experiments performed in order to assess the influence of word sense disambiguation on text classification. Three different corpora of online documents were disambiguated manually, and both original and disambiguated data were classified using various classification systems. The results do not support the hypothesis of WSD helping text classification very strongly, however some interesting general tendencies can be observed.