American politics in 3D: measuring multidimensional issue alignment in social media using social graphs and text data

Applied Network Science

Table 2 Groups of pairs of labeled users (according to criteria of Table 1), naming of the mined dimension perpendicular to the decision boundary of a multivariate logistic regression classification model, and the accuracy of the fitted model

Label 1	Label 2	Dimension	Precision	Recall	F1
Democrat	Republican	\(d_{Dem-Rep}\)	0.814	0.823	0.819
Trump sup.	Biden sup.	\(d_{Trump-Biden}\)	0.944	0.874	0.908
Liberal	Conservative	\(d_{Ideology}\)	0.877	0.848	0.862
Right	Left	\(d_{left-right}\)	0.615	0.484	0.542
Pro-BLM	Anti-BLM	\(d_{BLM}\)	0.488	0.205	0.288
BlackLM	BlueLM	\(d_{BlueLives}\)	0.687	0.925	0.789
Urban	Rural	\(d_{Urban}\)	0.569	0.655	0.609
Christian	Atheist	\(d_{Christian}\)	0.706	0.559	0.624
Pro-Gun	Anti-Guns	\(d_{Pro-Gun}\)	0.713	0.544	0.617
Pro-Communism	Anti-Communism	\(d_{Comm}\)	0.648	0.645	0.647
Liberal Life-Style	Anti-Lib. Life-Sty.	\(d_{LifeStyle}\)	0.537	0.867	0.664
Libertarian	Welfare	\(d_{Welfare}\)	0.695	0.720	0.707
Pro-Police	Anti-Police	\(d_{Police}\)	0.617	0.624	0.620
Pro-Military	Anti-Military	\(d_{Military}\)	0.599	0.474	0.529
Patriot	Anti-imperialism	\(d_{Patriot}\)	0.709	0.692	0.700
Global focus	Local focus	\(d_{Local}\)	0.547	0.676	0.605
DeepState	Rationalism	\(d_{Deepstate}\)	0.569	0.667	0.614
Anticapitalism	Entrepreneur	\(d_{Entrepreneur}\)	0.679	0.549	0.607
Justice	Liberty	\(d_{Liberty}\)	0.664	0.652	0.658
Pro-Incl. pronouns	Anti-incl. pronouns	\(d_{Anti-Pronouns}\)	0.630	0.710	0.667
Internationalization	Anti-globalization	\(d_{International}\)	0.731	0.583	0.649
Individual Views	Institutional Views	\(d_{Individual}\)	0.606	0.572	0.588
Pro-Mask	Anti-Mask	\(d_{Pro-Mask}\)	0.702	0.748	0.724
Science	Anti-Experts	\(d_{Science}\)	0.673	0.627	0.649