Abstract:

As an important preprocessing step in data mining and machine learning, feature selection has been gradually developed. The highdimensional characteristics of text data always declines the performance of categorization. Hence, feature selection can be employed as a dimensionreduction measure. With the fast evolution of classification methods and refinement of categories, multiclass text categorization gives rise to more challenges for feature selection. In this paper, we present a survey of the main problems and the stateofart feature selection methods, following the development track. Finally, we conclude the whole paper and give some future directions of research.

Key words: feature selection;text categorization;data mining;hierarchical structure