Emotion Detection with Transformers: A Comparative Study
This work addresses emotion detection in text for applications like sentiment analysis, but it is incremental as it compares existing transformer models without introducing new methods.
The study applied transformer-based models to emotion classification on text data, finding that common preprocessing techniques like removing punctuation and stop words can hinder performance, potentially because they disrupt contextual understanding.
In this study, we explore the application of transformer-based models for emotion classification on text data. We train and evaluate several pre-trained transformer models, on the Emotion dataset using different variants of transformers. The paper also analyzes some factors that in-fluence the performance of the model, such as the fine-tuning of the transformer layer, the trainability of the layer, and the preprocessing of the text data. Our analysis reveals that commonly applied techniques like removing punctuation and stop words can hinder model performance. This might be because transformers strength lies in understanding contextual relationships within text. Elements like punctuation and stop words can still convey sentiment or emphasis and removing them might disrupt this context.