A Chinese Text Classification Algorithm Based on Granular Computing

Classification is a basis ingredient of knowledge processing, and algorithms play a key role in a classifying system. In the paper, granular computing is applied in the domain of Chinese text classification. First, some concepts such as information granules, feature granules, decision granules, operations of granules and closeness of granules, are defined. Secondly, based on granular computing generating algorithm for building classification model and automatic classifying algorithm for Chinese text classification are proposed. Finally, they are illustrated by a real world example. It is shown that proposed algorithms are useful and effective. Because of granular computing aiming at reducing difficulties of problem solving, Chinese text classification based on granular computing is of theoretical meaning and practical value