问题
I'm attempting to find an example that demonstrates a Lucene or some other kind of index that can check an English first & last name combination for possible duplicates. The duplicate check needs to be able to take into account common nicknames, i.e. Bob for Robert and Bill for William, as well as spelling mistakes. Does anyone know of an example?
I plan to perform the duplicates search during user registration. The new user record needs to be checked against an index that has been built from the database table that stores the user names.
回答1:
I would use a SynonymFilter on the firstName while indexing, so that you have all possible combinations (Bob -> Robert, Robert -> Bob, etc...). Index the existing users you have.
Then use the QueryParser (without the SynonymFilter in the analyzer) to ask some fuzzy queries.
This is the code I came up with:
public class NameDuplicateTests {
private Analyzer analyzer;
private IndexSearcher searcher;
private IndexReader reader;
private QueryParser qp;
private final static Multimap<String, String> firstNameSynonyms;
static {
firstNameSynonyms = HashMultimap.create();
List<String> robertSynonyms = ImmutableList.of("Bob", "Bobby", "Robert");
for (String name: robertSynonyms) {
firstNameSynonyms.putAll(name, robertSynonyms);
}
List<String> willSynonyms = ImmutableList.of("William", "Will", "Bill", "Billy");
for (String name: willSynonyms) {
firstNameSynonyms.putAll(name, willSynonyms);
}
}
public static Analyzer createAnalyzer() {
return new Analyzer() {
@Override
public TokenStream tokenStream(String fieldName, Reader reader) {
TokenStream tokenizer = new WhitespaceTokenizer(reader);
if (fieldName.equals("firstName")) {
tokenizer = new SynonymFilter(tokenizer, new SynonymEngine() {
@Override
public String[] getSynonyms(String s) throws IOException {
return firstNameSynonyms.get(s).toArray(new String[0]);
}
});
}
return tokenizer;
}
};
}
@Before
public void setUp() throws Exception {
Directory dir = new RAMDirectory();
analyzer = createAnalyzer();
IndexWriter writer = new IndexWriter(dir, analyzer, IndexWriter.MaxFieldLength.UNLIMITED);
ImmutableList<String> firstNames = ImmutableList.of("William", "Robert", "Bobby", "Will", "Anton");
ImmutableList<String> lastNames = ImmutableList.of("Robert", "Williams", "Mayor", "Bob", "FunkyMother");
for (int id = 0; id < firstNames.size(); id++) {
Document doc = new Document();
doc.add(new Field("id", String.valueOf(id), Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("firstName", firstNames.get(id), Field.Store.YES, Field.Index.ANALYZED));
doc.add(new Field("lastName", lastNames.get(id), Field.Store.YES, Field.Index.NOT_ANALYZED));
writer.addDocument(doc);
}
writer.close();
qp = new QueryParser(Version.LUCENE_30, "firstName", new WhitespaceAnalyzer());
searcher = new IndexSearcher(dir);
reader = searcher.getIndexReader();
}
@After
public void tearDown() throws Exception {
searcher.close();
}
@Test
public void testNameFilter() throws Exception {
search("+firstName:Bob +lastName:Williams");
search("+firstName:Bob +lastName:Wolliam~");
}
private void search(String query) throws ParseException, IOException {
Query q = qp.parse(query);
System.out.println(q);
TopDocs res = searcher.search(q, 3);
for (ScoreDoc sd: res.scoreDocs) {
Document doc = reader.document(sd.doc);
System.out.println("Found " + doc.get("firstName") + " " + doc.get("lastName"));
}
}
}
Which results in:
+firstName:Bob +lastName:Williams
Found Robert Williams
+firstName:Bob +lastName:wolliam~0.5
Found Robert Williams
Hope that helps!
来源:https://stackoverflow.com/questions/8814190/detect-duplicate-english-names