如何在 Lucene 中查询自动完成/建议? [英] How to do query auto-completion/suggestions in Lucene?
问题描述
我正在寻找一种在 Lucene 中进行查询自动完成/建议的方法.我在谷歌上搜索了一下，玩了一会儿，但是我看到的所有示例似乎都是在 Solr 中设置过滤器.我们不使用 Solr，也不打算在不久的将来转向使用 Solr，而且 Solr 显然只是围绕 Lucene 进行了包装，所以我想一定有办法做到这一点！
I'm looking for a way to do query auto-completion/suggestions in Lucene. I've Googled around a bit and played around a bit, but all of the examples I've seen seem to be setting up filters in Solr. We don't use Solr and aren't planning to move to using Solr in the near future, and Solr is obviously just wrapping around Lucene anyway, so I imagine there must be a way to do it!
我研究过使用 EdgeNGramFilter，我意识到我必须在索引字段上运行过滤器并取出标记，然后将它们与输入的查询进行比较......我只是在努力制作将两者之间的联系转化为一些代码，非常感谢帮助！
I've looked into using EdgeNGramFilter, and I realise that I'd have to run the filter on the index fields and get the tokens out and then compare them against the inputted Query... I'm just struggling to make the connection between the two into a bit of code, so help is much appreciated!
为了明确我在寻找什么(我意识到我并没有说得太清楚，抱歉) - 我正在寻找一种解决方案，在搜索术语时，它会返回建议查询列表.在搜索字段中输入inter"时，它会返回一个建议查询列表，例如internet"、international"等.
To be clear on what I'm looking for (I realised I wasn't being overly clear, sorry) - I'm looking for a solution where when searching for a term, it'd return a list of suggested queries. When typing 'inter' into the search field, it'll come back with a list of suggested queries, such as 'internet', 'international', etc.
推荐答案
基于@Alexandre Victoor 的回答，我写了一个基于 contrib 包中的 Lucene Spellchecker 的小类(并使用其中包含的 LuceneDictionary)我想要什么.
Based on @Alexandre Victoor's answer, I wrote a little class based on the Lucene Spellchecker in the contrib package (and using the LuceneDictionary included in it) that does exactly what I want.
这允许从具有单个字段的单个源索引重新索引，并提供术语建议.结果按原始索引中与该术语匹配的文档数排序，因此更受欢迎的术语首先出现.似乎工作得很好:)
This allows re-indexing from a single source index with a single field, and provides suggestions for terms. Results are sorted by the number of matching documents with that term in the original index, so more popular terms appear first. Seems to work pretty well :)
import java.io.IOException;
import java.io.Reader;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.ISOLatin1AccentFilter;
import org.apache.lucene.analysis.LowerCaseFilter;
import org.apache.lucene.analysis.StopFilter;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.ngram.EdgeNGramTokenFilter;
import org.apache.lucene.analysis.ngram.EdgeNGramTokenFilter.Side;
import org.apache.lucene.analysis.standard.StandardFilter;
import org.apache.lucene.analysis.standard.StandardTokenizer;
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.lucene.index.CorruptIndexException;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.index.Term;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.ScoreDoc;
import org.apache.lucene.search.Sort;
import org.apache.lucene.search.TermQuery;
import org.apache.lucene.search.TopDocs;
import org.apache.lucene.search.spell.LuceneDictionary;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.FSDirectory;
/**
* Search term auto-completer, works for single terms (so use on the last term
* of the query).
* <p>
* Returns more popular terms first.
*
* @author Mat Mannion, M.Mannion@warwick.ac.uk
*/
public final class Autocompleter {
private static final String GRAMMED_WORDS_FIELD = "words";
private static final String SOURCE_WORD_FIELD = "sourceWord";
private static final String COUNT_FIELD = "count";
private static final String[] ENGLISH_STOP_WORDS = {
"a", "an", "and", "are", "as", "at", "be", "but", "by",
"for", "i", "if", "in", "into", "is",
"no", "not", "of", "on", "or", "s", "such",
"t", "that", "the", "their", "then", "there", "these",
"they", "this", "to", "was", "will", "with"
};
private final Directory autoCompleteDirectory;
private IndexReader autoCompleteReader;
private IndexSearcher autoCompleteSearcher;
public Autocompleter(String autoCompleteDir) throws IOException {
this.autoCompleteDirectory = FSDirectory.getDirectory(autoCompleteDir,
null);
reOpenReader();
}
public List<String> suggestTermsFor(String term) throws IOException {
// get the top 5 terms for query
Query query = new TermQuery(new Term(GRAMMED_WORDS_FIELD, term));
Sort sort = new Sort(COUNT_FIELD, true);
TopDocs docs = autoCompleteSearcher.search(query, null, 5, sort);
List<String> suggestions = new ArrayList<String>();
for (ScoreDoc doc : docs.scoreDocs) {
suggestions.add(autoCompleteReader.document(doc.doc).get(
SOURCE_WORD_FIELD));
}
return suggestions;
}
@SuppressWarnings("unchecked")
public void reIndex(Directory sourceDirectory, String fieldToAutocomplete)
throws CorruptIndexException, IOException {
// build a dictionary (from the spell package)
IndexReader sourceReader = IndexReader.open(sourceDirectory);
LuceneDictionary dict = new LuceneDictionary(sourceReader,
fieldToAutocomplete);
// code from
// org.apache.lucene.search.spell.SpellChecker.indexDictionary(
// Dictionary)
IndexReader.unlock(autoCompleteDirectory);
// use a custom analyzer so we can do EdgeNGramFiltering
IndexWriter writer = new IndexWriter(autoCompleteDirectory,
new Analyzer() {
public TokenStream tokenStream(String fieldName,
Reader reader) {
TokenStream result = new StandardTokenizer(reader);
result = new StandardFilter(result);
result = new LowerCaseFilter(result);
result = new ISOLatin1AccentFilter(result);
result = new StopFilter(result,
ENGLISH_STOP_WORDS);
result = new EdgeNGramTokenFilter(
result, Side.FRONT,1, 20);
return result;
}
}, true);
writer.setMergeFactor(300);
writer.setMaxBufferedDocs(150);
// go through every word, storing the original word (incl. n-grams)
// and the number of times it occurs
Map<String, Integer> wordsMap = new HashMap<String, Integer>();
Iterator<String> iter = (Iterator<String>) dict.getWordsIterator();
while (iter.hasNext()) {
String word = iter.next();
int len = word.length();
if (len < 3) {
continue; // too short we bail but "too long" is fine...
}
if (wordsMap.containsKey(word)) {
throw new IllegalStateException(
"This should never happen in Lucene 2.3.2");
// wordsMap.put(word, wordsMap.get(word) + 1);
} else {
// use the number of documents this word appears in
wordsMap.put(word, sourceReader.docFreq(new Term(
fieldToAutocomplete, word)));
}
}
for (String word : wordsMap.keySet()) {
// ok index the word
Document doc = new Document();
doc.add(new Field(SOURCE_WORD_FIELD, word, Field.Store.YES,
Field.Index.UN_TOKENIZED)); // orig term
doc.add(new Field(GRAMMED_WORDS_FIELD, word, Field.Store.YES,
Field.Index.TOKENIZED)); // grammed
doc.add(new Field(COUNT_FIELD,
Integer.toString(wordsMap.get(word)), Field.Store.NO,
Field.Index.UN_TOKENIZED)); // count
writer.addDocument(doc);
}
sourceReader.close();
// close writer
writer.optimize();
writer.close();
// re-open our reader
reOpenReader();
}
private void reOpenReader() throws CorruptIndexException, IOException {
if (autoCompleteReader == null) {
autoCompleteReader = IndexReader.open(autoCompleteDirectory);
} else {
autoCompleteReader.reopen();
}
autoCompleteSearcher = new IndexSearcher(autoCompleteReader);
}
public static void main(String[] args) throws Exception {
Autocompleter autocomplete = new Autocompleter("/index/autocomplete");
// run this to re-index from the current index, shouldn't need to do
// this very often
// autocomplete.reIndex(FSDirectory.getDirectory("/index/live", null),
// "content");
String term = "steve";
System.out.println(autocomplete.suggestTermsFor(term));
// prints [steve, steven, stevens, stevenson, stevenage]
}
}
这篇关于如何在 Lucene 中查询自动完成/建议?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
