Application of integrated gradients explainability to sociopsychological semantic markers
Researchers have applied the integrated gradient (IG) method to analyze sociopsychological semantic markers in text, moving beyond simple sentiment analysis. This technique reveals which specific words contribute to classification, enhancing explainability and providing deeper textual insights. The study focuses on markers like agency, using a verified deep learning classifier called BERTAgent, and explores training IG with limited datasets to identify salient words for different classes. AI
IMPACT Enhances AI model explainability and provides deeper insights into text analysis for sociopsychological markers.