在训练集中数据和标签已知的情况下，输入测试数据，将测试数据的特征与训练集中对应的特征进行相互比较，找到训练集中与之最为相似的前K个数据，则该测试数据对应的类别就是K个数据中出现次数最多的那个分类。(In the case where the training data and the tag are known, the test data is input, the characteristics of the test data are compared with the character