lengthmatcher.java

A Java port of LDC s Champollion sentence aligner (http://champollion.sourceforge.net). Intended aud

package net.sourceforge.akerblad;

public class LengthMatcher {

// #while(<>){
// # chomp;
// # if (/^(\d+) (\d+)$/) {
// # print match_sentences_len($1,$2,0.522,1.77),"\n";
// # }
// #}


    private static double pnorm(double z) {
        double t = 1/(1+0.2316419*z);
        double pd = 1-0.3989423*Math.exp(-z*z/2)*
    	((((1.330274429*t-1.821255978)*t
    	   +1.781477937)*t-0.356563782)*t+0.319381530)*t;

        return pd;
    }

    public static double matchSentencesLen(int len1, int len2, int yCharsPerXchar, int varPerXChar) {
        if (len1==0 && len2==0) {
    		return 0;
        }

        double mean = (len1+len2/yCharsPerXchar)/2;
        double z = (yCharsPerXchar*len1-len2)/Math.sqrt(varPerXChar*mean);

        if (z<0) {
            z = -z;
        }

        double pd = 2*(1-pnorm(z));
        return pd;
        //#return 0.5+1-pnorm($z);
    }
}