cat.lump.aq.textextraction.wikipedia.experiments

Class MeanTFxCategory

java.lang.Object

cat.lump.aq.textextraction.wikipedia.experiments.MeanTFxCategory

public class MeanTFxCategory
extends java.lang.Object

Wikiparable: Experiment 1 for evaluation. For a given category the most frequent 100 stems in the vocabulary (extracted from the root articles) are selected. The frequency of these stems is extracted from the collection and normalised according to different criteria. The result could be interpreted as the "density of in-domain terms". The collection can be taken as a whole or one can limit its size to mimic the size of the same collection in another model

Since:

May 29, 2015 TODO This class shares at lot of code with experiment 2 CorrelationsxCategory.java. Maybe they should extend a common class.

Author:

cristina

Constructor Summary

Constructors

Constructor and Description
MeanTFxCategory(java.lang.String language, java.lang.String model, java.lang.String modelRef, java.lang.String path)

Method Summary

Methods

Modifier and Type	Method and Description
static void	main(java.lang.String[] args) Main function to run the class, serves as example

Methods inherited from class java.lang.Object

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

Constructor Detail

MeanTFxCategory

public MeanTFxCategory(java.lang.String language,
               java.lang.String model,
               java.lang.String modelRef,
               java.lang.String path)

Method Detail

main

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

Main function to run the class, serves as example

Parameters:

args - -l Language of the articles -m Model to analyse -p Path to the root folder for the given language -c Category to analyse (optional)