Data Augmentation for Emotion Detection in Small Imbalanced Text Data

Emotion recognition in text, the task of identifying emotions such as joy or anger, is a challenging problem in NLP with many applications. One of the challenges is the shortage of available datasets that have been annotated with emotions. Certain existing datasets are small, follow different emotion taxonomies and display imbalance in their emotion distribution. In this work, we studied the impact of data augmentation techniques precisely when applied to small imbalanced datasets, for which current state-of-the-art models (such as RoBERTa) under-perform. Specifically, we utilized four data augmentation methods (Easy Data Augmentation EDA, static and contextual Embedding-based, and ProtAugment) on three datasets that come from different sources and vary in size, emotion categories and distributions. Our experimental results show that using the augmented data when training the classifier model leads to significant improvements. Finally, we conducted two case studies: a) directly using the popular chat-GPT API to paraphrase text using different prompts, and b) using external data to augment the training set. Results show the promising potential of these methods.

Deep Learning for Opinion Mining and Topic Classification of Course Reviews

Student opinions for a course are important to educators and administrators, regardless of the type of the course or the institution. Reading and manually analyzing open-ended feedback becomes infeasible for massive volumes of comments at institution level or online forums. In this paper, we collected and pre-processed a large number of course reviews publicly available online. We applied machine learning techniques with the goal to gain insight into student sentiments and topics. Specifically, we utilized current Natural Language Processing (NLP) techniques, such as word embeddings and deep neural networks, and state-of-the-art BERT (Bidirectional Encoder Representations from Transformers), RoBERTa (Robustly optimized BERT approach) and XLNet (Generalized Auto-regression Pre-training). We performed extensive experimentation to compare these techniques versus traditional approaches. This comparative study demonstrates how to apply modern machine learning approaches for sentiment polarity extraction and topic-based classification utilizing course feedback. For sentiment polarity, the top model was RoBERTa with 95.5\% accuracy and 84.7\% F1-macro, while for topic classification, an SVM (Support Vector Machine) was the top classifier with 79.8\% accuracy and 80.6\% F1-macro. We also provided an in-depth exploration of the effect of certain hyperparameters on the model performance and discussed our observations. These findings can be used by institutions and course providers as a guide for analyzing their own course feedback using NLP models towards self-evaluation and improvement.

Automatically Classifying Emotions based on Text: A Comparative Exploration of Different Datasets

Emotion Classification based on text is a task with many applications which has received growing interest in recent years. This paper presents a preliminary study with the goal to help researchers and practitioners gain insight into relatively new datasets as well as emotion classification in general. We focus on three datasets that were recently presented in the related literature, and we explore the performance of traditional as well as state-of-the-art deep learning models in the presence of different characteristics in the data. We also explore the use of data augmentation in order to improve performance. Our experimental work shows that state-of-the-art models such as RoBERTa perform the best for all cases. We also provide observations and discussion that highlight the complexity of emotion classification in these datasets and test out the applicability of the models to actual social media posts we collected and labeled.