This code is intended for the extraction of entity-centric network topics
as described in the publication (for details, see the provided link):

Andreas Spitz and Michael Gertz.
"Entity-centric Topic Extraction and Exploration: A Network-based Approach"
ECIR, Grenoble, France, March 26-29, 2018
https://dbs.ifi.uni-heidelberg.de/resources/nwtopics/

The contained script files can be executed in the R software environment
available at https://www.r-project.org/ by placing the scripts in the
same folder as the network data and calling (from within R):

source("r0x_nameOfFile.r")

The files build on each other and should be executed in order of ascending integers.
Note that the extraction of LDA topics requires the article contents, which we do not
include. You may have to first download the individual article contents from the article
URLs that are included in our data.

FILE CONTENTS:

r01_transformDataToBinary.r
Read the provided entity network data and transforms it to binary R Data files
for easier access in subsequent scripts.

r02_generateGraphs.r
Creates a set of new topic graphs from the provided data

r03_edgeTopicGrowthAndGraphExtraction.r
Implements the topic growth algorithm and saves extracted graphs to file

r04_extractTopicTermsFromGraphs.r
Extracts traditional topics (i.e., ranked lists of words) from the network

r05_articleContentDataPreparationLDA.r
Prepares a set of article contents (contained in plain text files) for
the extraction of LDA topics.

r06_topicExtractionLDA.r
Extracts LDA topics

r07_evalTopicModelsAndPlot.r
Performs a coverage comparison of network-based topics and LDA topics.
