jlattman/EU_speech_parliament_english

EU speech detection model

This model is trained to classify statements on the European Union. It finetuned a bert-base-german-cased model on 1700 sentences from the German parliament. It is trained to detect explicit and implicit mentionings of the European Union.

Code Book If a sentence mentions the EU in an explicit or implicit way, the model categorizes it as EU speech. Examples can be mentioning about policies, politicians, events, elections on the European level.

Examples: 1) With regard to the European Council, this means that a very precise distinction must be made as to where there are issues that we can no longer or can no longer fully address nationally.

1. 1.despite this adaptation to European law, the special features of the special features of German antitrust law must not be not be thrown overboard.

1. 1.but this is something we must do if we have an interest - and the vast majority of this House repeatedly emphasizes this - in using resources more efficiently in Europe through cooperation and in building a common industry in various areas.

Model Details Finetuned from model: google-bert/bert-based-cased Epochs: 3 Accuracy: 0.967

Lattmann, J. (2025, March 17). Detecting EU sentiment in texts: A LLM Machine Learning application for Euroscepticism research. https://doi.org/10.31219/osf.io/mravb_v1