是否有任何词干可用于印度语

是否有任何印度语语言的词干提取器的实现,如(印地语,泰卢固语)可用....     
已邀请:
印地语分析仪,带有干扰器,可在Lucene购买。它基于该算法(pdf)。     
hindi_stemmer是Ananthakrishnan Ramanathan和Durgesh D Rao在“A Lightweight Stemmer for Hindi”中描述的印地语词干的Python实现。     
我们创建了一个java版本的原始代码Python Hindi Stemmer 虽然原始印地语词干的作者使用变量L的方式还不是很清楚,但是有一个完整的代码可行 import java.util.ArrayList; import org.apache.commons.lang.StringUtils;
public class SimpleHindiStemmer {

/* This is not coming while pasting the code better copy from Python Code as given in suffixes[1]  */
private static String [] stem1 = new String [] { "ो", "े", "ू", "ु", "ी", "ि", "ा" };

/* This is not coming while pasting the code better copy from Python Code as given in suffixes[2]  */
private static String [] stem2 = new String [] { "कर", "ाओ", "िए", "ाई", "ाए", "ने", "नी", "ना", "ते", "ीं", "ती", "ता", "ाँ", "ां", "ों", "ें" };

private static String [] stem3 = new String [] { "ाकर", "ाइए", "ाईं", "ाया", "ेगी", "ेगा", "ोगी", "ोगे", "ाने", "ाना", "ाते", "ाती", "ाता", "तीं", "ाओं", "ाएं", "ुओं", "ुएं", "ुआं" };

private static String [] stem4 = new String [] { "ाएगी", "ाएगा", "ाओगी", "ाओगे", "एंगी", "ेंगी", "एंगे", "ेंगे", "ूंगी", "ूंगा", "ातीं", "नाओं", "नाएं", "ताओं", "ताएं", "ियाँ", "ियों", "ियां" };

private static String [] stem5 = new String [] { "ाएंगी", "ाएंगे", "ाऊंगी", "ाऊंगा", "ाइयाँ", "ाइयों", "ाइयां" };

private static ArrayList<String []> stemList = new ArrayList<String []>(5);

static int [] cut = new int [] { 0, 1, 1, 1, 2, 2 };

static {

    stemList.add(stem5);
    stemList.add(stem4);
    stemList.add(stem3);
    stemList.add(stem2);
    stemList.add(stem1);

}

public SimpleHindiStemmer(){

}

public static void main(String [] argv){

    SimpleHindiStemmer sm = new SimpleHindiStemmer();
    String word = "रास्ते";
    System.out.println(sm.stemprocess(word));
}

public String stemprocess(String word){


    int wlen = word.length();
    int wordlen = wlen*3;
    int icnt = 5;
    for (String [] stemwords : stemList){
        if(wordlen > (icnt + 1)){
            for(String sw: stemwords){
                if(StringUtils.endsWith(word, sw)){
                    return StringUtils.substring(word, 0, wlen - cut[icnt]);

                }
            }

        }


        icnt--;
    }

    return word;
}
}
如你所见,在某些情况下,utf-8字符未被很好地捕获。查看原始python代码,然后从那里复制后缀值。     
import java.util.Map;
import java.util.WeakHashMap;
/ **  *印地语光词干 - 从名词和形容词中删除数字,性别和案例后缀
public class HindiStemmerLight{

/**
 * A cache of words and their stems
 */
static private Map<String, String> cache = new WeakHashMap<String, String>();

/**
 * A buffer of the current word being stemmed
 */
private StringBuilder sb = new StringBuilder();

/**
 * Default constructor
 */
public HindiStemmerLight() {
}

public String stem(String word) {
    String result = cache.get(word);

    if (result != null)
        return result;

    // 
    sb.delete(0, sb.length());

    // 
    sb.append(word);

    /* remove the case endings from nouns and adjectives */
    remove_suffix(sb);



    result = sb.toString();
    cache.put(word, result);

    return result;
}

private void remove_suffix(StringBuilder word) {
    int len = word.length() - 1;

    /* article */

    if (len > 4) {
        if (word.substring( len- 2, len+1).equals("िया")) {
            word.delete(len-2 , len + 1);
            return;
        }
        if (word.substring( len- 2, len+1).equals("ियो")) {
            word.delete(len-2 , len + 1);
            return;
        }

    } /* end if len >4 */
    if (len > 3) {
        if (word.substring(len-1, len+1).equals("ाए")) {
            word.delete(len - 1, len + 1);
            return;
        }
        if (word.substring(len-1, len+1).equals(" ाओ")) {
            word.delete(len - 1, len + 1);
            return;
        }
        if (word.substring(len-1, len+1).equals(" ुआ")) {
            word.delete(len - 1, len + 1);
            return;
        }
        if (word.substring(len-1, len+1).equals(" ुओ")) {
            word.delete(len - 1, len + 1);
            return;
        }
        if (word.substring( len- 1, len+1).equals("ये")) {
            word.delete(len-1 , len + 1);
            return;
        }
        if (word.substring(len-1, len+1).equals(" ेन")) {
            word.delete(len - 1, len + 1);
            return;
        }
        if (word.substring(len-1, len+1).equals(" ेण")) {
            word.delete(len - 1, len + 1);
            return;
        }
        if (word.substring( len- 1, len+1).equals(" ीय")) {
            word.delete(len-1 , len + 1);
            return;
        }
        if (word.substring(len-1, len+1).equals("टी")) {
            word.delete(len - 1, len + 1);
            return;
        }
        if (word.substring(len-1, len+1).equals("ार")) {
            word.delete(len - 1, len + 1);
            return;
        }
        if (word.substring(len-1, len+1).equals("ाई")) {
            word.delete(len - 1, len + 1);
            return;
        }

    } /* end if len > 3 */
    if (len > 2) {
        if (word.substring(len, len+1).equals(" ा")) {
            word.delete(len , len + 1);
            return;
        }
        if (word.substring(len, len+1).equals(" े")) {
            word.delete(len , len + 1);
            return;
        }
        if (word.substring(len, len+1).equals(" ी")) {
            word.delete(len , len + 1);
            return;
        }
        if (word.substring(len, len+1).equals(" ो")) {
            word.delete(len , len + 1);
            return;
        }
        if (word.substring(len, len+1).equals("ि ")) {
            word.delete(len , len + 1);
            return;
        }
        if (word.substring(len, len+1).equals("अ")) {
            word.delete(len , len + 1);
            return;
        }

    } /* end if len > 2 */
    return;
}
}     

要回复问题请先登录注册