One text classification algorithm basing on tolerance rough set

At present, researches on theories and methods of data mining and knowledge discovery are in the ascendant, while rough set theory, as an effective way to deal with incomplete, inaccurate data, is attracting many researchers´ focus on its successful application in various fields. This paper first introduces the basic knowledge of rough set theory. Combining with the characteristics of text classification and properties of rough set, it then proposes approach of rough set based on tolerance and processes attribute reduction algorithm on it according to the importance and distribution of text properties in different texts and categories. By the experimental test, it achieves good results enriching the application of rough set and promoting the further development of text classification techniques.