Examples with structures._ParentDoc structures._ParentDoc used on opensource projects

Example 86 with structures._ParentDoc

use of structures._ParentDoc in project IR_Base by Linda-sunshine.

the class corrLDA_Gibbs method sampleInParentDoc.

protected void sampleInParentDoc(_Doc d) {
    _ParentDoc pDoc = (_ParentDoc) d;
    int wid, tid;
    double normalizedProb;
    for (_Word w : pDoc.getWords()) {
        wid = w.getIndex();
        tid = w.getTopic();
        pDoc.m_sstat[tid]--;
        if (m_collectCorpusStats) {
            word_topic_sstat[tid][wid]--;
            m_sstat[tid]--;
        }
        normalizedProb = 0;
        for (tid = 0; tid < number_of_topics; tid++) {
            double pWordTopic = parentWordByTopicProb(tid, wid);
            double pTopicPDoc = parentTopicInDocProb(tid, pDoc);
            double pTopicCDoc = parentChildInfluenceProb(tid, pDoc);
            m_topicProbCache[tid] = pWordTopic * pTopicPDoc * pTopicCDoc;
            normalizedProb += m_topicProbCache[tid];
        }
        normalizedProb *= m_rand.nextDouble();
        for (tid = 0; tid < number_of_topics; tid++) {
            normalizedProb -= m_topicProbCache[tid];
            if (normalizedProb < 0)
                break;
        }
        if (tid == number_of_topics)
            tid--;
        w.setTopic(tid);
        pDoc.m_sstat[tid]++;
        if (m_collectCorpusStats) {
            word_topic_sstat[tid][wid]++;
            m_sstat[tid]++;
        }
    }
}

Example 87 with structures._ParentDoc

use of structures._ParentDoc in project IR_Base by Linda-sunshine.

the class languageModelBaseLine method printTopChild4Stn.

protected void printTopChild4Stn(String filePrefix) {
    String topChild4StnFile = filePrefix + "/topChild4Stn.txt";
    try {
        PrintWriter pw = new PrintWriter(new File(topChild4StnFile));
        for (_Doc d : m_corpus.getCollection()) {
            if (d instanceof _ParentDoc) {
                _ParentDoc pDoc = (_ParentDoc) d;
                pw.println(pDoc.getName() + "\t" + pDoc.getSenetenceSize());
                for (_Stn stnObj : pDoc.getSentences()) {
                    // HashMap<String, Double> likelihoodMap = rankChild4StnByLikelihood(stnObj, pDoc);
                    HashMap<String, Double> likelihoodMap = rankChild4StnByLanguageModel(stnObj, pDoc);
                    // int i=0;
                    pw.print((stnObj.getIndex() + 1) + "\t");
                    for (Map.Entry<String, Double> e : sortHashMap4String(likelihoodMap, true)) {
                        // if(i==topK)
                        // break;
                        pw.print(e.getKey());
                        pw.print(":" + e.getValue());
                        pw.print("\t");
                    // i++;
                    }
                    pw.println();
                }
            }
        }
        pw.flush();
        pw.close();
    } catch (Exception e) {
        e.printStackTrace();
    }
}

Example 88 with structures._ParentDoc

use of structures._ParentDoc in project IR_Base by Linda-sunshine.

the class languageModelBaseLine method rankChild4StnByLikelihood.

protected HashMap<String, Double> rankChild4StnByLikelihood(_Stn stnObj, _ParentDoc pDoc) {
    HashMap<String, Double> childLikelihoodMap = new HashMap<String, Double>();
    for (_ChildDoc cDoc : pDoc.m_childDocs) {
        int cDocLen = cDoc.getTotalDocLength();
        _SparseFeature[] fv = cDoc.getSparse();
        double stnLogLikelihood = 0;
        double alphaDoc = m_smoothingMu / (m_smoothingMu + cDocLen);
        _SparseFeature[] sv = stnObj.getFv();
        for (_SparseFeature svWord : sv) {
            double featureLikelihood = 0;
            int wid = svWord.getIndex();
            double stnVal = svWord.getValue();
            int featureIndex = Utils.indexOf(fv, wid);
            if (featureIndex == -1)
                continue;
            double docVal = fv[featureIndex].getValue();
            double smoothingProb = docVal / (m_smoothingMu + cDocLen);
            smoothingProb += m_smoothingMu * m_wordSstat.get(wid) / (m_smoothingMu + cDocLen);
            featureLikelihood = Math.log(smoothingProb / (alphaDoc * m_wordSstat.get(wid)));
            stnLogLikelihood += stnVal * featureLikelihood;
        }
        stnLogLikelihood += stnObj.getLength() * Math.log(alphaDoc);
        childLikelihoodMap.put(cDoc.getName(), stnLogLikelihood);
    }
    return childLikelihoodMap;
}