Evaluation-Metrics

Sarcasm Detection with Transformers: When Perfect Accuracy Isn't Perfect

Learn how dataset bias can lead to misleading results in NLP: a sarcasm detection model that actually learned to classify news sources.