Java 使用Lucene使多值字段评分了解字段值计数
我正在尝试将(多单词)标记列表与每个文档关联。 因此,对于每个文档,我将添加多个StringField条目,并将“tag”作为fieldName 搜索时,我希望分数与我成功匹配的标记的比率成比例,例如:Java 使用Lucene使多值字段评分了解字段值计数,java,solr,lucene,Java,Solr,Lucene,我正在尝试将(多单词)标记列表与每个文档关联。 因此,对于每个文档,我将添加多个StringField条目,并将“tag”作为fieldName 搜索时,我希望分数与我成功匹配的标记的比率成比例,例如: 如果我匹配一半的标签,则为0.5 1.0,如果我将它们全部匹配 但似乎分数中没有考虑标签的数量 在对这四个文档进行测试时: - tags.put("doc1", "piano, electric guitar, violon"); - tags.put("doc2", "piano,
- 如果我匹配一半的标签,则为0.5
- 1.0,如果我将它们全部匹配
- tags.put("doc1", "piano, electric guitar, violon");
- tags.put("doc2", "piano, electric guitar");
- tags.put("doc3", "piano");
- tags.put("doc4", "electric guitar");
我得到的是:
- Score : 1.0
Doc : Document<stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<id:doc4> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:electric guitar>>
- Score : 1.0
Doc : Document<stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<id:doc2> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:piano> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:electric guitar>>
- Score : 1.0
Doc : Document<stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<id:doc1> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:piano> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:electric guitar> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:violon>>
-分数:1.0
文件:文件
-分数:1.0
文件:文件
-分数:1.0
文件:文件
我怎样才能改变这种行为?我是否错过了正确的做事方式
下面是我的测试代码
致以最良好的祝愿
雷诺
public class LuceneQueryTest {
Analyzer analyzer;
BasicIndex basicIndex;
LinkedList<String> phrases;
Query query;
Map<Document, Float> results;
@Test
public void testListOfTags() throws Exception {
analyzer = new StandardAnalyzer();
basicIndex = new BasicIndex(analyzer);
Map<String, String> tags = new HashMap();
tags.put("doc1", "piano, electric guitar, violon");
tags.put("doc2", "piano, electric guitar");
tags.put("doc3", "piano");
tags.put("doc4", "electric guitar");
Queue<String> queue = new LinkedList<>();
queue.addAll(tags.keySet());
basicIndex.index(new Supplier<Document>() {
public Document get() {
Document doc = new Document();
if (queue.isEmpty()) {
return null;
}
String docName = queue.poll();
System.out.println("**** "+docName);
String tag = tags.get(docName);
doc.add(new StringField("id", docName, Field.Store.YES));
for (String tagItem : tag.split("\\,")) {
System.out.println(tagItem);
Field tagField;
tagField = new StringField("tag",tagItem,Field.Store.YES);
System.out.println(tagField);
doc.add(tagField);
}
return doc;
}
});
BooleanQuery booleanQuery = new BooleanQuery();
//booleanQuery.add(new TermQuery(new Term("tag", "piano")), BooleanClause.Occur.SHOULD);
booleanQuery.add(new TermQuery(new Term("tag", "electric guitar")), BooleanClause.Occur.SHOULD);
//Query parsedQuery = new QueryParser("tag", analyzer).parse("tag:\"electric guitar\"");
query = booleanQuery;
//query = parsedQuery;
System.out.println(query);
results = basicIndex.search(query);
displayResults(results);
System.out.println(Arrays.toString(basicIndex.document(3).getValues("tag")));
}
private void displayResults(Map<Document, Float> results) {
results.forEach((Document doc, Float score) -> {
System.out.println("Score : " + score + " \n Doc : " + doc);
});
}
}
公共类LuceneQueryTest{
分析仪;
BasicIndex BasicIndex;
链接列表短语;
查询;
地图结果;
@试验
public void testListOfTags()引发异常{
analyzer=新的StandardAnalyzer();
basicIndex=新的basicIndex(分析仪);
Map tags=newhashmap();
放置(“doc1”,“钢琴,电吉他,小提琴”);
标签。放置(“doc2”,“钢琴,电吉他”);
标签。放置(“doc3”,“钢琴”);
标签。放置(“doc4”,“电吉他”);
Queue Queue=new LinkedList();
queue.addAll(tags.keySet());
basicIndex.索引(新供应商(){
公共文档get(){
单据单据=新单据();
if(queue.isEmpty()){
返回null;
}
字符串docName=queue.poll();
System.out.println(“*****”+docName);
String tag=tags.get(docName);
doc.add(新的StringField(“id”,docName,Field.Store.YES));
用于(字符串tagItem:tag.split(“\\,”)){
系统输出打印项次(标记项);
字段标记字段;
tagField=new-StringField(“tag”,tagItem,Field.Store.YES);
System.out.println(标记字段);
文件添加(标记字段);
}
退货单;
}
});
BooleanQuery BooleanQuery=新的BooleanQuery();
//add(newtermquery(newterm(“tag”,“piano”)),BooleanClause.occure.SHOULD);
add(新术语查询(新术语(“标记”、“电吉他”))、BooleanClause.occure.SHOULD);
//Query parsedQuery=new QueryParser(“tag”,analyzer).parse(“tag:\“电吉他”);
查询=布尔查询;
//query=parsedQuery;
System.out.println(查询);
结果=basicIndex.search(查询);
显示结果(结果);
System.out.println(Arrays.toString(basicIndex.document(3.getValues)(“tag”));
}
私有void显示结果(映射结果){
结果.forEach((文档文档,浮动分数)->{
System.out.println(“分数:+Score+”\n文档:+Doc);
});
}
}
BasicIndex(测试实用程序)类的代码:
import java.io.IOException;
导入java.util.LinkedHashMap;
导入java.util.LinkedList;
导入java.util.Map;
导入java.util.function.function;
导入java.util.function.Supplier;
导入org.apache.lucene.analysis.Analyzer;
导入org.apache.lucene.document.document;
导入org.apache.lucene.document.Field;
导入org.apache.lucene.document.TextField;
导入org.apache.lucene.index.DirectoryReader;
导入org.apache.lucene.index.IndexReader;
导入org.apache.lucene.index.IndexWriter;
导入org.apache.lucene.index.IndexWriterConfig;
导入org.apache.lucene.search.indexsearch;
导入org.apache.lucene.search.Query;
导入org.apache.lucene.search.ScoreDoc;
导入org.apache.lucene.search.TopScoreDocCollector;
导入org.apache.lucene.store.Directory;
导入org.apache.lucene.store.RAMDirectory;
导入org.apache.lucene.util.Version;
/**
*
*@作者雷诺
*/
公共类BasicIndex{
final Directory=new RAMDirectory();
最终索引编写器索引编写器;
最终分析仪;
公用BasicIndex(分析仪){
this.analyzer=分析器;
this.indexWriter=newIndexWriter();
}
公共分析器getAnalyzer(){
回波分析仪;
}
私有索引编写器newIndexWriter(){
IndexWriterConfig配置=新的IndexWriterConfig(最新版本,analyzer);
试一试{
返回新的IndexWriter(目录,配置);
}捕获(IOEX异常){
抛出新的运行时异常(ex);
}
}
公共索引搜索器newindexsearch(){
返回新的IndexSearcher(newIndexReader());
}
公共索引阅读器newIndexReader(){
索引阅读器;
试一试{
reader=DirectoryReader.open(目录);
}捕获(IOEX异常){
抛出ExceptionUtils.asRuntimeException(ex);
}
返回读取器;
}
公共无效索引(LinkedList词组,最终字符串字段名){
索引(短语,(字符串短语)->{
单据单据=新单据();
Field workField=新文本字段(字段名、短语、字段.Store.YES);
单据新增(工作字段);
退货单;
});
}
公共作废索引(供应商文件){
文件;
而((document=documents.get())!=null){
试一试{
indexWriter.addDocument(文档);
}捕获(IOE异常){
抛出ExceptionUtils.asRuntimeException(e);
}
}
close();
}
公共无效索引(LinkedList短语、函数docBuilder){
for(字符串短语:短语){
试一试{
indexWriter.addDocument(docBuilder.apply(短语));
}捕获(IOE异常){
抛出ExceptionUtils.asRuntimeException(e);
}
}
close();
}
私人作废关闭(){
IOUtils.closes(indexWriter);
}
公共地图搜索(查询){
final IndexSearcher IndexSearcher=newIndexSearcher();
int hitsPerPage=10;
上部核心
import java.io.IOException;
import java.util.LinkedHashMap;
import java.util.LinkedList;
import java.util.Map;
import java.util.function.Function;
import java.util.function.Supplier;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.lucene.document.TextField;
import org.apache.lucene.index.DirectoryReader;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.index.IndexWriterConfig;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.ScoreDoc;
import org.apache.lucene.search.TopScoreDocCollector;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.RAMDirectory;
import org.apache.lucene.util.Version;
/**
*
* @author renaud
*/
public class BasicIndex {
final Directory directory = new RAMDirectory();
final IndexWriter indexWriter;
final Analyzer analyzer;
public BasicIndex(Analyzer analyzer) {
this.analyzer = analyzer;
this.indexWriter = newIndexWriter();
}
public Analyzer getAnalyzer() {
return analyzer;
}
private IndexWriter newIndexWriter() {
IndexWriterConfig config = new IndexWriterConfig(Version.LATEST, analyzer);
try {
return new IndexWriter(directory, config);
} catch (IOException ex) {
throw new RuntimeException(ex);
}
}
public IndexSearcher newIndexSearcher() {
return new IndexSearcher(newIndexReader());
}
public IndexReader newIndexReader() {
IndexReader reader;
try {
reader = DirectoryReader.open(directory);
} catch (IOException ex) {
throw ExceptionUtils.asRuntimeException(ex);
}
return reader;
}
public void index(LinkedList<String> phrases, final String fieldName) {
index(phrases, (String phrase) -> {
Document doc = new Document();
Field workField = new TextField(fieldName, phrase, Field.Store.YES);
doc.add(workField);
return doc;
});
}
public void index(Supplier<Document> documents) {
Document document;
while ((document = documents.get()) != null) {
try {
indexWriter.addDocument(document);
} catch (IOException e) {
throw ExceptionUtils.asRuntimeException(e);
}
}
close();
}
public void index(LinkedList<String> phrases, Function<String, Document> docBuilder) {
for (String phrase : phrases) {
try {
indexWriter.addDocument(docBuilder.apply(phrase));
} catch (IOException e) {
throw ExceptionUtils.asRuntimeException(e);
}
}
close();
}
private void close() {
IOUtils.closeSilently(indexWriter);
}
public Map<Document, Float> search(Query query) {
final IndexSearcher indexSearcher = newIndexSearcher();
int hitsPerPage = 10;
TopScoreDocCollector collector = TopScoreDocCollector.create(hitsPerPage, true);
try {
indexSearcher.search(query, collector);
} catch (IOException ex) {
throw new RuntimeException(ex);
}
ScoreDoc[] hits = collector.topDocs().scoreDocs;
Map<Document, Float> results = new LinkedHashMap<>();
for (int i = 0; i < hits.length; ++i) {
ScoreDoc scoreDoc = hits[i];
int docId = scoreDoc.doc;
float score = scoreDoc.score;
Document doc;
try {
doc = indexSearcher.doc(docId);
} catch (IOException ex) {
throw new RuntimeException(ex);
}
results.put(doc, score);
}
return results;
}
public Document document(int i){
try {
return newIndexSearcher().doc(i);
} catch (IOException ex) {
throw new RuntimeException(ex);
}
}
Score : 0.5
Doc : Document<stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<id:doc4> stored<count:1> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:electric guitar>>
Score : 0.33333334
Doc : Document<stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<id:doc2> stored<count:2> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:piano> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:electric guitar>>
Score : 0.25
Doc : Document<stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<id:doc1> stored<count:3> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:piano> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:electric guitar> stored,indexed,tokenized,omitNorms,indexOptions=DOCS_ONLY<tag:violon>>
@Test
public void testListOfTags() throws Exception {
analyzer = new StandardAnalyzer();
basicIndex = new BasicIndex(analyzer);
Map<String, String> tags = new HashMap();
tags.put("doc1", "piano, electric guitar, violon");
tags.put("doc2", "piano, electric guitar");
tags.put("doc3", "piano");
tags.put("doc4", "electric guitar");
Queue<String> queue = new LinkedList<>();
queue.addAll(tags.keySet());
basicIndex.index(new Supplier<Document>() {
public Document get() {
Document doc = new Document();
if (queue.isEmpty()) {
return null;
}
String docName = queue.poll();
System.out.println("**** " + docName);
String tag = tags.get(docName);
doc.add(new StringField("id", docName, Field.Store.YES));
String[] tags = tag.split("\\,");
Field tagCountField = new IntField("count", tags.length, Field.Store.YES);
doc.add(tagCountField);
for (String tagItem : tags) {
System.out.println(tagItem);
Field tagField;
tagField = new StringField("tag", tagItem.trim(), Field.Store.YES);
System.out.println(tagField);
doc.add(tagField);
}
return doc;
}
});
BooleanQuery booleanQuery = new BooleanQuery();
//booleanQuery.add(new TermQuery(new Term("tag", "piano")), BooleanClause.Occur.SHOULD);
booleanQuery.add(new TermQuery(new Term("tag", "electric guitar")), BooleanClause.Occur.SHOULD);
//Query parsedQuery = new QueryParser("tag", analyzer).parse("tag:\"electric guitar\"");
query = booleanQuery;
//query = parsedQuery;
ValueSource boostSource = new ReciprocalFloatFunction(new IntFieldSource("count"), 1, 1, 1);
query = new BoostedQuery(query, boostSource);
System.out.println(query);
results = basicIndex.search(query);
displayResults(results);
System.out.println(Arrays.toString(basicIndex.document(3).getValues("tag")));
}