How Populist are Parties? Measuring Degrees of Populism in Party Manifestos Using Supervised Machine Learning

Cocco, Jessica Di; Monechi, Bernardo

doi:10.1017/pan.2021.29

Cited by 23 publications

(19 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In supervised machine learning, still scarcely applied to measure aspects of populism (see Di Cocco and Monechi, 2022), the researcher in contrast 'trains' an algorithm, simulating coders' activity (for more information on this 'training', see appendix table C) 13 . In order to train the algorithm, we derived emotional scores per sentence using a Random Forest (RF) classification algorithm (Breiman, 2001) 14 , adopting a bag-of-words approach after text pre-processing.…”

Section: Methods and Sourcesmentioning

confidence: 99%

Populism and emotions: a comparative study using Machine Learning

Caiani

Cocco

2023

Riv. ital. sci. polit.

Self Cite

View full text Add to dashboard Cite

This study aims to unpack the mobilization of emotions in the political discourse of populist and non-populist parties and above all, across ‘varieties of populism’ (right wing vs. left wing or hybrid). Is there an empirical connection between emotions and populism? Are all types of populisms alike with regards to the emotional appeals within their political discourse? Focusing on Italy as a crucial case for populist communication and using a novel methodological approach based on supervised machine learning, it systematically investigates the intensity and trends of specific emotions in political discourses (institutional and informal, i.e. leaders’ speeches) of all Italian political parties over the last 20 years, for a corpus of more than 13,000 sentences analysed. The findings confirm that (i) populists tend to use more (and a broader repertoire of) emotional appeals than non-populist parties; however (ii) overall, there is an increase in the use of these appeals in the Italian political party discourse over time, especially in terms of negative emotions; and, most importantly, (iii) different types of emotions are mobilized by different types of populisms. Right wing populism mainly uses negative emotions while left wing or hybrid populism employs positive emotional appeals. The communication arena (party manifestoes vs. speeches) nevertheless does matter in the degree and types of emotions mobilized by political actors. This study identifies important implications for research on emotional appeals in politics, populist communication and political campaigning, and populist contagion from an emotion-based perspective.

show abstract

Section: Methods and Sourcesmentioning

confidence: 99%

Populism and emotions: a comparative study using Machine Learning

Caiani

Cocco

2023

Riv. ital. sci. polit.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Lastly, we have shown that the fine-tuned models are competitive in terms of training time and inference time. Future research could explore the broader application of pretrained language models, alongside cross-domain classifiers, to other research questions, such as populism prediction (Cocco and Monechi 2022), sentiment and stance analysis (Bestvater and Monroe 2022), and party position analysis (Herrmann and Döring 2021), as well as the optimization of pretrained language models in training and inference.…”

Section: Training and Inference Timementioning

confidence: 99%

Topic Classification for Political Texts with Pretrained Language Models

Wang¹

2023

Polit. Anal.

View full text Add to dashboard Cite

Supervised topic classification requires labeled data. This often becomes a bottleneck as high-quality labeled data are expensive to acquire. To overcome the data scarcity problem, scholars have recently proposed to use cross-domain topic classification to take advantage of preexisting labeled datasets. Cross-domain topic classification only requires limited annotation in the target domain to verify its cross-domain accuracy. In this letter, we propose supervised topic classification with pretrained language models as an alternative. We show that language models fine-tuned with 70% of the small annotated dataset in the target corpus could outperform models trained using large cross-domain datasets by 27% and that models fine-tuned with 10% of the annotated dataset could already outperform the cross-domain classifiers. Our models are competitive in terms of training time and inference time. Researchers interested in supervised learning with limited labeled data should find our results useful. Our code and data are publicly available.1

show abstract

“…As attention to populist parties has grown, so too has the methodological work on identifying populism in text, including in party manifestos (Rooduijn and A key challenge has been to identify populism in short text, such as a sentence or a paragraph, in order to estimate the degree or amount of populism in a document. Given that no dataset exists that labels populist speech at the sentence level, recent work has proposed training sentence-level supervised classifiers to identify populism using manifesto-level labels (Di Cocco and Monechi 2021). This approach has been criticized, for, among other things, for relying on document-level labels to train a sentence-level classifier when most sentences in a populist party's manifesto will not be recognizably populist (Jankowski and Huber 2023).…”

Section: Application 3: Synthetic Data For Zero-shot Classifierstrain...mentioning

confidence: 99%

Synthetically generated text for supervised text analysis

Halterman¹

2023

Preprint

View full text Add to dashboard Cite

Supervised text models are a valuable tool for political scientists but present several obstacles to their use, including the expense of hand-labeling documents, the difficulty of retrieving rare relevant documents for annotation, and copyright and privacy concerns involved in sharing annotated documents. This article proposes a partial solution to these three issues, in the form of controlled generation of synthetic text with large language models. I provide a conceptual overview of text generation, guidance on when researchers should prefer different techniques for generating synthetic text, a discussion of ethics, and a simple technique for improving the quality of synthetic text. I demonstrate the usefulness of synthetic text with three applications: generating synthetic tweets describing the fighting in Ukraine, synthetic news articles describing specified political events for training an event detection system, and a multilingual corpus of populist manifesto statements for training a sentence-level populism classifier.

show abstract

How Populist are Parties? Measuring Degrees of Populism in Party Manifestos Using Supervised Machine Learning

Cited by 23 publications

References 59 publications

Populism and emotions: a comparative study using Machine Learning

Populism and emotions: a comparative study using Machine Learning

Topic Classification for Political Texts with Pretrained Language Models

Synthetically generated text for supervised text analysis

Contact Info

Product

Resources

About