基于三支决策的KNN渐进式文本分类方法

在传统的文本分类中，KNN算法以其简单、分类准确率高、非参数得到了广泛的应用。但是传统KNN算法在进行文本分类的过程中，需要计算待分类文本与每一个训练样本的相似度，当面对海量的文本时，分类的效果会明显降低。针对此问题，提出了一种基于三支决策的KNN渐进式文本分类方法用于提高其分类效率，结合三支决策在分类问题中的优势，将三支决策与KNN算法相结合，对标题、摘要、关键词等进行渐进式的分类处理，从而完成待分类文本的分类，提高文本分类的效率和性能。实验表明，该算法能够在确保KNN算法分类准确率的基础上，同时提高分类效率。

Progressive text classification based on three-way decision and KNN algorithm

In the traditional text classification, KNN algorithm is widely used in text classification because of its simplicity, high classification accuracy and non parameter. However, in the process of text classification, traditional KNN needs to calculate the similarity between the text to be classified and each training sample. When faced with massive text, the classification effect will be significantly reduced. To solve this problem, this paper proposed a KNN progressive text classification method based on three-way decision making to improve its classification efficiency. This algorithm combined with the advantages of three-way decision in classification, and combined with KNN algorithm, through the title, abstract, keywords and other progressive classification processing, so as to complete the text classification, improve the efficiency and performance of text classification. Experiments show that the algorithm can ensure the classification accuracy of KNN algorithm and improve the classification efficiency at the same time.