Included is supplementary material for the paper "Knowledge Production: Analysing Gender and Country Dependent Factors in Research Topics through Term Communities" by P. Bakshi-Hamm and A. Hamm

================================================================================

Folder structure:
	Counts: contains diagrams depicting the country dependent gender distribution within seven sub area corpora in absolute and relative terms
	Topics: contains term communities and their interpretation and distribution within the documents for seven sub area corpora
	AbsShares: contains absolute topic contributions within seven sub area corpora

================================================================================

Details:

	File naming conventions in folder Counts:
		COUNT_<sub_area_name>_country_rel_stacked.png: stacked column bar chart of the shares of author gender classes per country for the named subject sub area, like in Fig. 1 of the paper
		COUNT_<sub_area_name>_country_rel_stacked.png: stacked column bar chart of the absolute document counts in author gender classes per country for the named subject sub area

	The folder Topics contains one sub folder per subject sub area with the following naming convention:
		S<sub_area_code>: Folder with topics of the sub area with the given Scopus code

	Each of these sub folders contains the following files:
		Topic_<n>.png: stratified word cloud for the term community number n, like Fig. 2 and Fig. 3 of the paper
		labels.txt: labels for the topic interpretation of the term communities
		tsne.html: 2-dimensional tSNE visualisation of how documents form topic clusters. Each dot represents a document, with the colour depending on the dominant topic in the document
		details.xlsx: 
			Sheet "Abs" contains the topic term counts per document. The column "key" gives the identifier of each document, the column "title" its title, the column "year" its publication year, the column "countries" a list of countries in which the document originates, and the column "authorgender_95" the author gender class to which the document belongs.
			Sheet "Rel" is similar to the sheet "Abs" but contains the fractional shares of each topic per document. 
			Sheets "Topic_<n>" contains the topic terms of term community number n. Semantically related terms are in the same column - with decreasing specificity from top to bottom. 

	File naming conventions in the folder AbsShares:
		<sub_area_name>_abs_abc.png: absolute topic contributions for the named subject sub area, grouped by a) publication year, b) country, c) authors' gender, like Fig. 11 of the paper. The vertical axis has a logarithmic scale. 
        