我在 weka 中编写了一些代码,它应用了一个过滤器,然后应用了一个聚类算法,但是我得到了以下错误:
线程“主”weka.core.WekaException 中的异常:weka.clusterers.Cobweb:没有属性!
代码:
这有什么问题??:
StringToWordVector stringToWordVector = new StringToWordVector();
// bigrams to 4-grams
NGramTokenizer nGramTokenizer = new NGramTokenizer();
nGramTokenizer.setNGramMinSize(2);
nGramTokenizer.setNGramMinSize(4);
stringToWordVector.setTokenizer(nGramTokenizer);
// TF-IDF
stringToWordVector.setTFTransform(true);
stringToWordVector.setIDFTransform(true);
stringToWordVector.setUseStoplist(true);
stringToWordVector.setLowerCaseTokens(true);
ArffLoader loader = new ArffLoader();
loader.setFile(new File("./learningRelation.arff"));
Instances structure = loader.getStructure();
stringToWordVector.setInputFormat(structure);
structure = Filter.useFilter(structure, stringToWordVector);
structure.setClassIndex(TestInstances.NO_CLASS);
Cobweb cw = new Cobweb();
cw.buildClusterer(structure);
Instance current;
while ((current = loader.getNextInstance(structure)) != null)
cw.updateClusterer(current);
cw.updateFinished();
// output generated model
System.out.println(cw);
文件 learningRelation.arff 是这样的:
@relation 推文
@attribute 推文字符串
@数据
'社交商务 - 介绍 | 通过 ronsela http://t.co/LrGHQKrmOQ社交电子商务营销
'社交商务 - 介绍 | 通过 ronsela http://t.co/LrGHQKrmOQ社交电子商务营销