/** 人名消歧,比如.邓颖超生前->邓颖 超生 前 fix to 丁颖超 生 前! 规则的方式增加如果两个人名之间连接是- ， ·，•则连接 */
public static void nameAmbiguity(Term[] terms) {
Term from = null;
Term term = null;
Term next = null;
for (int i = 0; i < terms.length - 1; i++) {
term = terms[i];
if (term != null && term.termNatures() == TermNatures.NR && term.getName().length() == 2) {
next = terms[i + 2];
if (next.termNatures().personAttr.split > 0) {
term.setName(term.getName() + next.getName().charAt(0));
terms[i + 2] = null;
terms[i + 3] = new Term(next.getName().substring(1), next.getOffe(), TermNatures.NW);
TermUtil.termLink(term, terms[i + 3]);
TermUtil.termLink(terms[i + 3], next.to());
}
}
}
// 外国人名修正
for (int i = 0; i < terms.length; i++) {
term = terms[i];
if (term != null
&& term.getName().length() == 1
&& i > 0
&& WordAlert.CharCover(term.getName().charAt(0)) == '·') {
from = term.from();
next = term.to();
if (from.natrue().natureStr.startsWith("nr") && next.natrue().natureStr.startsWith("nr")) {
from.setName(from.getName() + term.getName() + next.getName());
TermUtil.termLink(from, next.to());
terms[i] = null;
terms[i + 1] = null;
}
}
}
}
/**
* 从一个词的词性到另一个词的词的分数
*
* @param form 前面的词
* @param to 后面的词
* @return 分数
*/
public static double compuScore(Term from, Term to) {
double frequency = from.termNatures().allFreq + 1;
if (frequency < 0) {
double score = from.score() + MAX_FREQUENCE;
from.score(score);
return score;
}
int nTwoWordsFreq = NgramLibrary.getTwoWordFreq(from, to);
double value =
-Math.log(
dSmoothingPara * frequency / (MAX_FREQUENCE + 80000)
+ (1 - dSmoothingPara) * ((1 - dTemp) * nTwoWordsFreq / frequency + dTemp));
if (value < 0) {
value += frequency;
}
return from.score() + value;
}
/**
* 词性词频词长.计算出来一个分数
*
* @param from
* @param term
* @return
*/
public static double compuScoreFreq(Term from, Term term) {
// TODO Auto-generated method stub
return from.termNatures().allFreq + term.termNatures().allFreq;
}
