adding builed russian suffix evresitcs
git-svn-id: https://russianmorphology.googlecode.com/svn/trunk@4 d817d54c-26ab-11de-abc9-2f7d1455ff7a
This commit is contained in:
@ -13,12 +13,12 @@ import java.util.*;
|
||||
import java.util.concurrent.atomic.AtomicInteger;
|
||||
|
||||
|
||||
public class SuffixResearcher {
|
||||
public class EvristicBuilder {
|
||||
public static void main(String[] args) throws IOException {
|
||||
IgnoredFormReader formReader = new IgnoredFormReader("igoredFrom.txt");
|
||||
IgnoredFormReader formReader = new IgnoredFormReader("data/igoredFrom.txt");
|
||||
Set<String> form = formReader.getIngnoredFroms();
|
||||
|
||||
DictonaryReader dictonaryReader = new DictonaryReader("morphs.mrd", form);
|
||||
DictonaryReader dictonaryReader = new DictonaryReader("dictonary/Dicts/SrcMorph/RusSrc/morphs.mrd", form);
|
||||
StatiticsCollectors statiticsCollectors = new StatiticsCollectors();
|
||||
dictonaryReader.proccess(statiticsCollectors);
|
||||
Collection<SuffixCounter> counterCollection = statiticsCollectors.getStatititics().values();
|
||||
@ -34,26 +34,6 @@ public class SuffixResearcher {
|
||||
evristic.addEvristic(((SuffixCounter) objects[i]).getSuffixEvristic());
|
||||
}
|
||||
|
||||
final AtomicInteger good = new AtomicInteger(0);
|
||||
final AtomicInteger bad = new AtomicInteger(0);
|
||||
final FileWriter writer = new FileWriter("incorret.txt");
|
||||
dictonaryReader.proccess(new WordProccessor(){
|
||||
public void proccess(WordCard wordCard) throws IOException {
|
||||
for(String wordForm:wordCard.getWordsFroms()){
|
||||
String cf = wordCard.getCanonicalFrom();
|
||||
if (evristic.getNormalForm(wordForm).equals(cf)){
|
||||
good.incrementAndGet();
|
||||
} else{
|
||||
writer.write(wordForm + " c " + cf + " f " + evristic.getNormalForm(wordForm) + "\n");
|
||||
bad.incrementAndGet();
|
||||
}
|
||||
}
|
||||
}
|
||||
});
|
||||
writer.close();
|
||||
|
||||
System.out.println("Good " + good + " Bad " + bad);
|
||||
|
||||
evristic.writeToFile("evriticsb");
|
||||
evristic.writeToFile("src/main/resources/russianSuffixesEvristics.txt");
|
||||
}
|
||||
}
|
694813
src/main/resources/russianSuffixesEvristics.txt
Normal file
694813
src/main/resources/russianSuffixesEvristics.txt
Normal file
File diff suppressed because it is too large
Load Diff
Reference in New Issue
Block a user