基于隶属度限幅特征VSM的文本分类模型

摘要

Through the expression of text characteristic based on fuzzy qualifier and then definition of fuzzy feature using the fuzzy function, represented the text as the imposed membership degree limiting text feature vector. Constructuring membership degree limiting class feature matrix, and mapping each group of texts belong to the same class to its class expectation vector. All of class expectation vectors constructed the membership degree limiting characteristic VSM. Based on that, presented a new text-classification model and the experiment shows that the model is efficient.%通过文档基于模糊限定词的特征表达,定义特征的模糊函数,将文档表示为隶属度限幅的特征向量,构造文本集隶属度限幅的类特征矩阵,将每一类文本集映射为类期望向量,所有类期望向量便构成了隶属度限幅的特征VSM.在此基础上设计了一种新的文本分类模型.实验结果证明,该分类模型能有效实现文本分类.