org.knowceans.corpus.analysis
Class LdaTopicVariationAnalyser

java.lang.Object
  org.knowceans.corpus.analysis.LdaTopicVariationAnalyser

public class LdaTopicVariationAnalyser
extends java.lang.Object

LdaTopicVariationAnalyser analyses the variation of the topic parameters through the document by comparing the theta vectors of the document to these of the sentences

TODO: not completed.

Author:

heinrich

Field Summary

private TermCorpusOld	tc
private double[][]	theta

Constructor Summary

LdaTopicVariationAnalyser()
does nothing (yet)

Method Summary

(package private) double	getDocumentStats(double[][] thetad, java.util.Vector<double[][]> thetas) calculate the distance statistics between the document topics and the sentence topics
private boolean	isSentence(java.lang.String string) check whether the string is a sentence name
static void	main(java.lang.String[] args)
double	run(java.lang.String docsfile, java.lang.String model, boolean zip)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

theta

private double[][] theta

tc

private TermCorpusOld tc

Constructor Detail

LdaTopicVariationAnalyser

public LdaTopicVariationAnalyser()

does nothing (yet)

Method Detail

main

public static void main(java.lang.String[] args)

run

public double run(java.lang.String docsfile,
                  java.lang.String model,
                  boolean zip)

isSentence

private boolean isSentence(java.lang.String string)

check whether the string is a sentence name

Parameters:

string -

Returns:

getDocumentStats

double getDocumentStats(double[][] thetad,
                        java.util.Vector<double[][]> thetas)

calculate the distance statistics between the document topics and the sentence topics

Parameters:

thetad -

thetas -

Returns: