Lucene的评分(score)机制研究
Posted 一天不进步,就是退步!
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了Lucene的评分(score)机制研究相关的知识,希望对你有一定的参考价值。
首先,需要学习Lucene的评分计算公式——
分值计算方式为查询语句q中每个项t与文档d的匹配分值之和,当然还有权重的因素。其中每一项的意思如下表所示:
表3.5 |
评分公式中的因子 |
评分因子 |
描 述 |
tf(t in d) |
项频率因子——文档(d)中出现项(t)的频率 |
idf(t) |
项在倒排文档中出现的频率:它被用来衡量项的“唯一”性.出现频率较高的term具有较低的idf,出现较少的term具有较高的idf |
boost(t.field in d) |
域和文档的加权,在索引期间设置.你可以用该方法 对某个域或文档进行静态单独加权 |
lengthNorm(t.field in d) |
域的归一化(Normalization)值,表示域中包含的项数量.该值在索引期间计算,并保存在索引norm中.对于该因子,更短的域(或更少的语汇单元)能获得更大的加权 |
coord(q,d) |
协调因子(Coordination factor),基于文档中包含查询的项个数.该因子会对包含更多搜索项的文档进行类似AND的加权 |
queryNorm(q) |
每个査询的归一化值,指毎个查询项权重的平方和
|
通过Searcher.explain(Query query, int doc)方法可以查看某个文档的得分的具体构成。 示例:
public class ScoreSortTest { public final static String INDEX_STORE_PATH = "index"; public static void main(String[] args) throws Exception { IndexWriter writer = new IndexWriter(INDEX_STORE_PATH, new StandardAnalyzer(), true); writer.setUseCompoundFile(false); Document doc1 = new Document(); Document doc2 = new Document(); Document doc3 = new Document(); Field f1 = new Field("bookname","bc bc", Field.Store.YES, Field.Index.TOKENIZED); Field f2 = new Field("bookname","ab bc", Field.Store.YES, Field.Index.TOKENIZED); Field f3 = new Field("bookname","ab bc cd", Field.Store.YES, Field.Index.TOKENIZED); doc1.add(f1); doc2.add(f2); doc3.add(f3); writer.addDocument(doc1); writer.addDocument(doc2); writer.addDocument(doc3); writer.close(); IndexSearcher searcher = new IndexSearcher(INDEX_STORE_PATH); TermQuery q = new TermQuery(new Term("bookname", "bc")); q.setBoost(2f); Hits hits = searcher.search(q); for(int i=0; i<hits.length();i++){ Document doc = hits.doc(i); System.out.print(doc.get("bookname") + "\t\t"); System.out.println(hits.score(i)); System.out.println(searcher.explain(q, hits.id(i)));// } } }
运行结果:
bc bc 0.629606 0.629606 = (MATCH) fieldWeight(bookname:bc in 0), product of: 1.4142135 = tf(termFreq(bookname:bc)=2) 0.71231794 = idf(docFreq=3, numDocs=3) 0.625 = fieldNorm(field=bookname, doc=0) ab bc 0.4451987 0.4451987 = (MATCH) fieldWeight(bookname:bc in 1), product of: 1.0 = tf(termFreq(bookname:bc)=1) 0.71231794 = idf(docFreq=3, numDocs=3) 0.625 = fieldNorm(field=bookname, doc=1) ab bc cd 0.35615897 0.35615897 = (MATCH) fieldWeight(bookname:bc in 2), product of: 1.0 = tf(termFreq(bookname:bc)=1) 0.71231794 = idf(docFreq=3, numDocs=3) 0.5 = fieldNorm(field=bookname, doc=2)
涉及到的源码:
idf的计算
idf是项在倒排文档中出现的频率,计算方式为
- /** Implemented as <code>log(numDocs/(docFreq+1)) + 1</code>. */
- @Override
- public float idf(long docFreq, long numDocs) {
- return (float)(Math.log(numDocs/(double)(docFreq+1)) + 1.0);
- }
docFreq是根据指定关键字进行检索,检索到的Document的数量,我们测试的docFreq=14;numDocs是指索引文件中总共的Document的数量,我们测试的numDocs=1453。用计算器验证一下,没有错误,这里就不啰嗦了。
queryNorm的计算
queryNorm的计算在DefaultSimilarity类中实现,如下所示:
- /** Implemented as <code>1/sqrt(sumOfSquaredWeights)</code>. */
- public float queryNorm(float sumOfSquaredWeights) {
- return (float)(1.0 / Math.sqrt(sumOfSquaredWeights));
- }
这里,sumOfSquaredWeights的计算是在org.apache.lucene.search.TermQuery.TermWeight类中的sumOfSquaredWeights方法实现:
- public float sumOfSquaredWeights() {
- queryWeight = idf * getBoost(); // compute query weight
- return queryWeight * queryWeight; // square it
- }
其实默认情况下,sumOfSquaredWeights = idf * idf,因为Lucune中默认的boost = 1.0。
fieldWeight的计算
在org/apache/lucene/search/similarities/TFIDFSimilarity.java的explainScore方法中有:
- // explain field weight
- Explanation fieldExpl = new Explanation();
- fieldExpl.setDescription("fieldWeight in "+doc+
- ", product of:");
- Explanation tfExplanation = new Explanation();
- tfExplanation.setValue(tf(freq.getValue()));
- tfExplanation.setDescription("tf(freq="+freq.getValue()+"), with freq of:");
- tfExplanation.addDetail(freq);
- fieldExpl.addDetail(tfExplanation);
- fieldExpl.addDetail(stats.idf);
- Explanation fieldNormExpl = new Explanation();
- float fieldNorm = norms != null ? decodeNormValue(norms.get(doc)) : 1.0f;
- fieldNormExpl.setValue(fieldNorm);
- fieldNormExpl.setDescription("fieldNorm(doc="+doc+")");
- fieldExpl.addDetail(fieldNormExpl);
- fieldExpl.setValue(tfExplanation.getValue() *
- stats.idf.getValue() *
- fieldNormExpl.getValue());
- result.addDetail(fieldExpl);
重点是这一句:
- fieldExpl.setValue(tfExplanation.getValue() *
- stats.idf.getValue() *
- fieldNormExpl.getValue());
使用计算式表示就是
fieldWeight = tf * idf * fieldNorm
tf和idf的计算参考前面的,fieldNorm的计算在索引的时候确定了,此时直接从索引文件中读取,这个方法并没有给出直接的计算。如果使用DefaultSimilarity的话,它实际上就是lengthNorm,域越长的话Norm越小,在org/apache/lucene/search/similarities/DefaultSimilarity.java里面有关于它的计算:
- public float lengthNorm(FieldInvertState state) {
- final int numTerms;
- if (discountOverlaps)
- numTerms = state.getLength() - state.getNumOverlap();
- else
- numTerms = state.getLength();
- return state.getBoost() * ((float) (1.0 / Math.sqrt(numTerms)));
- }
参考文献:
【1】http://www.hankcs.com/program/java/lucene-scoring-algorithm-explained.html
【2】http://grantbb.iteye.com/blog/181802
以上是关于Lucene的评分(score)机制研究的主要内容,如果未能解决你的问题,请参考以下文章