java - 如何在java中使用lucene添加自定义停用词

Question

我正在使用 lucene 删除英语停用词，但我的要求是删除英语停用词和自定义停用词。下面是我使用 lucene 删除英语停用词的代码。

我的示例代码：

public class Stopwords_remove {
    public String removeStopWords(String string) throws IOException 
    {
        StandardAnalyzer ana = new StandardAnalyzer(Version.LUCENE_30);
        TokenStream tokenStream = new StandardTokenizer(Version.LUCENE_36,newStringReader(string));
        StringBuilder sb = new StringBuilder();
        tokenStream = new StopFilter(Version.LUCENE_36, tokenStream, ana.STOP_WORDS_SET);
        CharTermAttribute token = tokenStream.getAttribute(CharTermAttribute.class);
        while (tokenStream.incrementToken()) 
        {
            if (sb.length() > 0) 
            {
                sb.append(" ");
            }
            sb.append(token.toString());
        }
        return sb.toString();
    }

    public static void main(String args[]) throws IOException
    {
          String text = "this is a java project written by james.";
          Stopwords_remove stopwords = new Stopwords_remove();
          stopwords.removeStopWords(text);

    }
}

输出：java project written james.

所需输出：java project james.

我怎样才能做到这一点？

Answer 1

您可以将额外的停用词添加到标准英语停用词集的副本中，或者只是添加另一个 StopFilter。喜欢：

TokenStream tokenStream = new StandardTokenizer(Version.LUCENE_36, new StringReader(string));
CharArraySet stopSet = CharArraySet.copy(Version.LUCENE_36, StandardAnalyzer.STOP_WORD_SET);
stopSet.add("add");
stopSet.add("your");
stopSet.add("stop");
stopSet.add("words");
tokenStream = new StopFilter(Version.LUCENE_36, tokenStream, stopSet);
//Or, if you just need the added stopwords in a standardanalyzer, you could just pass this stopfilter into the StandardAnalyzer...
//analyzer = new StandardAnalyzer(Version.LUCENE_36, stopSet);

或者：

TokenStream tokenStream = new StandardTokenizer(Version.LUCENE_36, new StringReader(string));
tokenStream = new StopFilter(Version.LUCENE_36, tokenStream, StandardAnalyzer.STOP_WORDS_SET);
List<String> stopWords = //your list of stop words.....
tokenStream = new StopFilter(Version.LUCENE_36, tokenStream, StopFilter.makeStopSet(Version.LUCENE_36, stopWords));

如果您正在尝试创建自己的分析器，则最好遵循类似于分析器文档中的示例的模式。