New ML models tackle political sentiment analysis on social media

Researchers have developed two machine learning approaches, one using XGBoost and another based on BERT, to tackle the challenge of multiclass sentiment analysis for identifying political viewpoints on social media. Both models were trained and evaluated on a labeled dataset of political social media posts. The XGBoost model achieved an F1-score of 0.2835, while the BERT-based model reached an F1-score of 0.2806, highlighting the difficulty in classifying complex political discourse. AI

IMPACT This research provides a baseline for multiclass political sentiment analysis, highlighting the challenges in classifying complex social media discourse.

RANK_REASON The cluster contains an academic paper detailing new machine learning models for sentiment analysis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• BERT
• DagsHub
• Girme Yohannis Bade
• Gotit.pub
• Hugging Face
• natural language processing
• ScienceCast
• XGBoost

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →