Diachronic Analysis of German Parliamentary Proceedings: Ideological Shifts through the Lens of Political Biases

TL;DR

This paper investigates historical political and racist biases in German parliamentary proceedings from 1867 to 2020 using diachronic semantic models and novel bias measurement techniques, revealing trends consistent with historical perceptions.

Contribution

It introduces a new corpus, DeuPARL, and a novel bias measurement method based on term co-occurrences and graph propagation for diachronic analysis.

Findings

Bias measurements align with known historical trends.

Semantic models effectively capture ideological shifts.

Method proves viable for analyzing historical biases.

Abstract

We analyze bias in historical corpora as encoded in diachronic distributional semantic models by focusing on two specific forms of bias, namely a political (i.e., anti-communism) and racist (i.e., antisemitism) one. For this, we use a new corpus of German parliamentary proceedings, DeuPARL, spanning the period 1867--2020. We complement this analysis of historical biases in diachronic word embeddings with a novel measure of bias on the basis of term co-occurrences and graph-based label propagation. The results of our bias measurements align with commonly perceived historical trends of antisemitic and anti-communist biases in German politics in different time periods, thus indicating the viability of analyzing historical bias trends using semantic spaces induced from historical corpora.