ThangDLU at #SMM4H 2024: Encoder-decoder models for classifying text data on social disorders in children and adolescents
This work addresses social media mining for health applications, specifically for researchers and practitioners analyzing mental health data, but it is incremental as it applies existing methods to new tasks.
The paper tackled classification of tweets on social disorders in children and adolescents, applying encoder-decoder models like BART-base and T5-small with data augmentation, achieving best F1 scores of 0.627 for multi-class classification on social anxiety and 0.841 for binary classification on medical disorders.
This paper describes our participation in Task 3 and Task 5 of the #SMM4H (Social Media Mining for Health) 2024 Workshop, explicitly targeting the classification challenges within tweet data. Task 3 is a multi-class classification task centered on tweets discussing the impact of outdoor environments on symptoms of social anxiety. Task 5 involves a binary classification task focusing on tweets reporting medical disorders in children. We applied transfer learning from pre-trained encoder-decoder models such as BART-base and T5-small to identify the labels of a set of given tweets. We also presented some data augmentation methods to see their impact on the model performance. Finally, the systems obtained the best F1 score of 0.627 in Task 3 and the best F1 score of 0.841 in Task 5.