在开源中文analysis,我选择了paoding analysis,link:http://code.google.com/p/paoding
配置如下:
在CLASSPATH 上面加入:E:\eclipse\paoding-analysis.properties
在用户变量加入:PAODING_DIC_HOME=》E:\dic
把dic文件,就是字词文本文件放到对应的目录。
测试代码:
- package test;
- import java.io.IOException;
- import java.io.Reader;
- import java.io.StringReader;
- import net.paoding.analysis.analyzer.PaodingAnalyzer;
- import net.paoding.analysis.analyzer.PaodingTokenizer;
- import org.apache.lucene.analysis.Analyzer;
- import org.apache.lucene.analysis.Token;
- public class Test1 {
- public static void main(String[] argv){
- Analyzer analyzer = new PaodingAnalyzer();
- String testString = "中华人民共和国";
- Reader r = new StringReader(testString);
- PaodingTokenizer ts = (PaodingTokenizer) analyzer.tokenStream("", r);
- Token t;
- try {
- while((t = ts.next()) != null){
- System.out.println(t);
- }
- } catch (IOException e) {
- e.printStackTrace();
- }
- }
- }
package test;
import java.io.IOException;
import java.io.Reader;
import java.io.StringReader;
import net.paoding.analysis.analyzer.PaodingAnalyzer;
import net.paoding.analysis.analyzer.PaodingTokenizer;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.Token;
public class Test1 {
public static void main(String[] argv){
Analyzer analyzer = new PaodingAnalyzer();
String testString = "中华人民共和国";
Reader r = new StringReader(testString);
PaodingTokenizer ts = (PaodingTokenizer) analyzer.tokenStream("", r);
Token t;
try {
while((t = ts.next()) != null){
System.out.println(t);
}
} catch (IOException e) {
e.printStackTrace();
}
}
}
结果:
- (中华,0,2)
- (华人,1,3)
- (人民,2,4)
- (共和,4,6)
- (共和国,4,7)
- log4j:WARN No appenders could be found for logger (net.paoding.analysis.knife.PaodingMaker).
- log4j:WARN Please initialize the log4j system properly.