Beyond Sentiment: Comparing Traditional NLP and LLM-Based Multi-Dimensional Analysis for Political News Evaluation
The study compares traditional RoBERTa sentiment analysis with an LLM-based framework to evaluate political news. Traditional models suffer from 'neutral collapse,' misclassifying rich political text as neutral. The LLM approach provides a richer, multi-dimensional analysis capturing framing, bias, and sensationalism, offering a more accurate tool for social science research.