Question

我正在尝试了解Lucene SpanNearQuery并编写了一个虚拟示例。我正在寻找“不”，其次是“狐狸”。我希望文档3作为唯一的命中返回。但是，我最终没有得到任何打击。任何关于我可能做错的想法都将受到赞赏。

以下是代码：

//索引

public void doSpanIndexing()  throws IOException {   

IndexWriter writer=new IndexWriter(directory, AnalyzerUtil.getPorterStemmerAnalyzer(new StandardAnalyzer(Version.LUCENE_30)),IndexWriter.MaxFieldLength.LIMITED);

 Document doc1=new Document();
 doc1.add(new Field("content", " brown fox jumped ", Field.Store.YES, Index.ANALYZED,  Field.TermVector.WITH_POSITIONS_OFFSETS));
 writer.addDocument(doc1);


 Document doc2=new Document();
 doc2.add(new Field("content", "foxes not jumped over the huge fence", Field.Store.YES, Index.ANALYZED,Field.TermVector.WITH_POSITIONS_OFFSETS));
 writer.addDocument(doc2);

 Document doc3=new Document();
 doc3.add(new Field("content", " brown not fox", Field.Store.YES, Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS));
 writer.addDocument(doc3);


}

//搜索
public void doSpanSearching（String text）抛出CorruptIndexException，IOException，ParseException {

 IndexSearcher searcher=new IndexSearcher(directory);

 SpanTermQuery term1 = new SpanTermQuery(new Term("content", "not"));
 SpanTermQuery term2 = new SpanTermQuery(new Term("content", text));
 SpanNearQuery query = new SpanNearQuery(new SpanQuery[] {term1, term2}, 5, true);
 TopDocs topDocs=searcher.search(query,5);

for(int i=0; i<topDocs.totalHits; i++) {
   System.out.println("Hit Document number: "+topDocs.scoreDocs[i].doc);
   System.out.println("Hit Document score: "+topDocs.scoreDocs[i].score);
   Document result=searcher.doc(topDocs.scoreDocs[i].doc);
   System.out.println("Search result "+(i+1)+ " is "+result.get("content"));

  }

}

Answer 1

“Not”是标准分析器中的停用词（即从文本中删除）。你能用另一个不是停止词的词来尝试吗？

Lucene SpanNearQuery

1 个答案: