关键字@relation 预期 weka java答案

【问题标题】：keyword @relation expected weka java关键字@relation 预期 weka java
【发布时间】：2023-03-17 07:59:01
【问题描述】：

我在java中使用weka实现了一个kmeans算法

所以我有这段 java 代码，我复制自：Cannot handle any class attribute! kmeans java

public class demo {
    public demo() throws Exception {
        // TODO Auto-generated constructor stub
        BufferedReader breader = null;
        breader = new BufferedReader(new FileReader(
                "D:/workspace_java/JenaTutorial/movie.arff"));
        Instances Train = new Instances(breader);
        //Train.setClassIndex(Train.numAttributes() - 1); // comment out this line
        SimpleKMeans kMeans = new SimpleKMeans();
        kMeans.setSeed(10);
        kMeans.setPreserveInstancesOrder(true);
        kMeans.setNumClusters(3);
        kMeans.buildClusterer(Train);
        int[] assignments = kMeans.getAssignments();
        int i = 0;
        for (int clusterNum : assignments) {
            System.out.printf("Instance %d -> Cluster %d", i, clusterNum);
            i++;
        }
        breader.close();
    }
    public static void main(String[] args) throws Exception {
        // TODO Auto-generated method stub
        new demo();
    }
}

它不是我的文件“movie.arff”的提取：

@attribute title string
@attribute annee string
@attribute genre string
@attribute genre1 string
@attribute genre2 string
@attribute genre3 string
@attribute actor string
@attribute actor1 string
@attribute actor2 string
@attribute actor3 string
@attribute actor4 string
@attribute actor5 string
@attribute actor6 string
@attribute actor7 string
@attribute actor8 string
@attribute actor9 string
@attribute writer string
@attribute writer1 string
@attribute writer2 string
@attribute writer3 string
@attribute writer4 string
@attribute writer5 string
@attribute writer6 string
@attribute writer7 string
@attribute writer8 string
@attribute writer9 string


@data
'Toy Story','1995','Adventure','Animation','Children','Comedy','Fantasy','Tom Hanks','Wallace Shawn','John Ratzenberger','John Morris ','Annie Potts','Don Rickles','Laurie Metcalf','Tim Allen','Jim Varney','Erik von Detten','Joel Cohen (writer)','Joss Whedon','Andrew Stanton','Alec Sokolow',

'Jumanji','1995','Adventure','Children','Fantasy','?','Robin Williams','Kirsten Dunst','David Alan Grier','Bonnie Hunt','Bebe Neuwirth','Jonathan Hyde','?','?','?','?','Greg Taylor (author)','Jonathan Hensleigh','?','?','?','?','?','?','?','?','Greg Taylor (author)','Jonathan Hensleigh','?','?',


'Grumpier Old Men','1995','Comedy','Romance','?','?','Sophia Loren','Walter Matthau','Jack Lemmon','Ann-Margret','Burgess Meredith','Daryl Hannah','Ann Morgan Guilbert','Kevin Pollak','?','?','Mark Steven Johnson','?','?','?','?','?','?','?','?','?','Mark Steven Johnson','?','?','?',


'Waiting to Exhale','1995','Comedy','Drama','Romance','?','Whitney Houston','Lela Rochon','Loretta Devine','Angela Bassett','?','?','?','?','?','?','Ronald Bass','Terry McMillan','?','?','?','?','?','?','?','?','Ronald Bass','Terry McMillan','?','?',


'Father of the Bride Part II','1995','Comedy','?','?','?','Steve Martin','Martin Short','Diane Keaton','?','?','?','?','?','?','?','Charles Shyer','Albert Hackett','Frances Goodrich','Nancy Meyers','?','?','?','?','?','?','Charles Shyer','Albert Hackett','Frances Goodrich','Nancy Meyers',

但在执行中我有这个例外：

Exception in thread "main" java.io.IOException: keyword @relation expected, read Token[@attribute], line 1
at weka.core.converters.ArffLoader$ArffReader.errorMessage(ArffLoader.java:354)
at weka.core.converters.ArffLoader$ArffReader.readHeader(ArffLoader.java:806)
at weka.core.converters.ArffLoader$ArffReader.<init>(ArffLoader.java:204)
at weka.core.Instances.<init>(Instances.java:135)
at wakaproject.demo.<init>(demo.java:14)
at wakaproject.demo.main(demo.java:31)

你能帮帮我吗！

【问题讨论】：

尝试在 weka explorer 中加载你的文件，你会得到同样的错误。修复你的 arff 文件，使其可以在 weka explorer 中加载，然后尝试编写 java 代码。
重新阅读 arff 格式规范。您的文件不是正确的 arff 文件。
也明白k-means只能使用数值数据。顾名思义，它需要计算 mean。 “玩具总动员”和“冒险”是什么意思？

标签： java exception attributes weka k-means

【解决方案1】：

我在使用 Java 程序时也遇到过类似的问题。

程序总是需要有效的 *.arff 文件格式，如下所示：

@relation **<file name>**
@attribute <attribute1-name> ...
@attribute <attribute2-name> ...

@data
1,1
1,2
1,1.333333
...
...

如果您在使用 CSV 文件时遇到类似的异常。请在此处查看https://www.researchgate.net/publication/280612446_Using_WEKA_in_your_java_code_Clustering

【讨论】：

【解决方案2】：

正如它所说，文件必须以@relation 而不是@attribute 开头

【讨论】：