A pipeline combining LDA, Top2Vec, GPT-2, similarity analysis, and human evaluation extracts policy agendas from social media with reported good inter-rater agreement and cosine similarity scores.
Tucker, and Richard Bonneau
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Using machine learning to build public policy agenda from social media conversations
A pipeline combining LDA, Top2Vec, GPT-2, similarity analysis, and human evaluation extracts policy agendas from social media with reported good inter-rater agreement and cosine similarity scores.