我尝试使用Weka的NaiveBayesUpdateable分类器。我的数据包含名义属性和数字属性:@relationcars@attributecountry{FR,UK,...}@attributecity{London,Paris,...}@attributecar_make{Toyota,BMW,...}@attributepricenumeric%%carprice@attributesalesnumeric%%numberofcarssold我需要根据其他属性预测销售额(数字!)。我知道我不能在Weka中使用数字属性进行贝叶斯分类。一种技术是将数字属性的值拆分为长度为k的N
在Integer.java中,有如下一段代码:/***AllpossiblecharsforrepresentinganumberasaString*/finalstaticchar[]digits={'0','1','2','3','4','5','6','7','8','9','a','b','c','d','e','f','g','h','i','j','k','l','m','n','o','p','q','r','s','t','u','v','w','x','y','z'};我认为您需要的所有数字/字符都在0-9和字母A到F范围内。只有在表示数字时才会使用字母(A、B、C、