35 lines
869 B
Java
35 lines
869 B
Java
package com.star.util;
|
|
|
|
import java.io.IOException;
|
|
import java.io.Reader;
|
|
import java.io.StringReader;
|
|
import java.util.LinkedList;
|
|
import java.util.List;
|
|
|
|
import org.wltea.analyzer.core.IKSegmenter;
|
|
import org.wltea.analyzer.core.Lexeme;
|
|
|
|
public class Extractor {
|
|
public static List<String> extractAtoms(String str){
|
|
//System.out.print(name + ">>>");
|
|
str = Normalizer.normalize(str);
|
|
//System.out.print(name + ">>>");
|
|
Reader text = new StringReader(str);
|
|
IKSegmenter iks = new IKSegmenter(text, true);
|
|
Lexeme term = null;
|
|
List<String> items = new LinkedList<String>();
|
|
try {
|
|
while((term = iks.next()) != null){
|
|
items.add(term.getLexemeText().toString());
|
|
}
|
|
} catch (IOException e) {
|
|
// TODO Auto-generated catch block
|
|
e.printStackTrace();
|
|
System.out.println("分词出现错误");
|
|
System.exit(1);
|
|
}
|
|
return items;
|
|
}
|
|
|
|
}
|