APSI (Automated Political Stance Identification) is a theory-driven text analysis tool that estimates political positions expressed in text using Natural Language Inference (NLI).

APSI analyses text across three dimensions: Economic Left–Right ideology, Support for Liberal Democratic values, and Populist vs. Pluralist rhetoric.

APSI is intended for research, journalism, think tanks, civil society organisations, and critical analysis of political discourse.

APSI is theory-driven, transparent, interpretable, reproducible, and scalable — designed to enable auditing of how scores are produced.

NLI is a method where a model estimates whether a text supports (entails) a hypothesis statement, rather than relying on keywords alone.

APSI does not train a supervised classifier on labelled stance datasets. Instead, it uses a hypothesis-based scoring framework with a pretrained political NLI model, enabling zero-shot and interpretable results.

Zero-shot means APSI can estimate political stances without training on a labelled stance dataset for your specific task.

Hypotheses are derived from political theory and expert survey codebooks, then written as natural-language statements suitable for NLI scoring.

APSI averages entailment probabilities for each side of a dimension, takes the difference, and scales it to a 0–10 score (clipped to stay within bounds).

APSI runs a topic relevance pre-check. If the text does not strongly match political topic hypotheses (threshold 0.6), it returns no score.

The tool returns score = NA and an explanation that it cannot infer a political stance from the text.

APSI uses the Political DEBATE model (Burnham et al. 2024), which is specialised for political discourse analysis using Natural Language Inference.

It is built on the DeBERTa V3 large architecture, which was first fine-tuned for general-purpose NLI classification and then further trained on the PolNLI political dataset.

PolNLI is a political Natural Language Inference dataset used to fine-tune the Political DEBATE model. It contains documents from social media, news outlets, congressional bills, court case summaries, and more.

The political fine-tuning supports multiple political NLP tasks: stance detection, hate speech detection, event extraction, and topic classification.

Lower scores indicate more economically left-leaning language; higher scores indicate more economically right-leaning language. The scale runs from 0 (Strong Left) to 10 (Strong Right).

Lower values indicate more pluralist rhetoric; higher values indicate more populist rhetoric. The tool maps score ranges to labels such as "Pluralist", "Moderate", and "Populist".

Lower scores indicate weaker support for liberal democratic values; higher scores indicate stronger support.

Confidence reflects the internal consistency of the model's hypothesis probabilities. Lower variance across hypotheses produces higher confidence.

Contradiction is flagged when both sides of a dimension receive comparatively strong support, suggesting mixed or conflicting signals in the text.

APSI outputs a score, an interpretation label, a confidence level, a contradiction status, and indicators of which hypotheses most influenced the score.

You can analyse speeches, articles, social media posts, policy statements, interviews, and transcripts — any text where political language might appear.

No. APSI analyses only the text provided and does not know the author, audience, or circumstances surrounding the text.

APSI measures rhetorical framing in language, not verified beliefs, intent, or factual truth.

APSI was evaluated by comparing its scores to political science expert ratings on a validation set of political texts (20 texts per dimension).

A total of 147 expert responses were collected across the three dimensions.

Yes. Outputs are approximate and can be inaccurate or incomplete due to complex political language, model limitations, or bias.

APSI depends on hypothesis design, can reflect training-data bias, measures rhetoric rather than intent, and lacks real-world context about the text or its author.

No. APSI should not be used for legal, employment, or automated decision-making affecting people. Human interpretation and critical judgment are required.

Users are responsible for the content they submit, for applying appropriate critical judgment to the results, and for ensuring their use of the tool complies with copyright and applicable legal requirements.