You Shall Know a Tool by the Traces it Leaves: The Predictability of Sentiment Analysis Tools
This highlights a critical reliability issue in sentiment analysis for NLP practitioners and researchers, warning against uncritical use of such tools.
The study found that sentiment analysis tools produce inconsistent results across different datasets and languages, and the specific tool used can be predicted from its output with an average F1-score of 0.89 for English corpora, revealing algorithmic bias.
If sentiment analysis tools were valid classifiers, one would expect them to provide comparable results for sentiment classification on different kinds of corpora and for different languages. In line with results of previous studies we show that sentiment analysis tools disagree on the same dataset. Going beyond previous studies we show that the sentiment tool used for sentiment annotation can even be predicted from its outcome, revealing an algorithmic bias of sentiment analysis. Based on Twitter, Wikipedia and different news corpora from the English, German and French languages, our classifiers separate sentiment tools with an averaged F1-score of 0.89 (for the English corpora). We therefore warn against taking sentiment annotations as face value and argue for the need of more and systematic NLP evaluation studies.