Investigating Semi-Supervised Learning Algorithms in Text Datasets
This work addresses the challenge of applying semi-supervised learning to text data, where augmentation methods are less effective than in images, but it is incremental as it evaluates existing algorithms without introducing new ones.
The study compared semi-supervised learning algorithms that do not require data augmentation, such as self-training and tri-training, on text datasets, finding that tri-training with disagreement performed closest to an Oracle but still had a performance gap.
Using large training datasets enhances the generalization capabilities of neural networks. Semi-supervised learning (SSL) is useful when there are few labeled data and a lot of unlabeled data. SSL methods that use data augmentation are most successful for image datasets. In contrast, texts do not have consistent augmentation methods as images. Consequently, methods that use augmentation are not as effective in text data as they are in image data. In this study, we compared SSL algorithms that do not require augmentation; these are self-training, co-training, tri-training, and tri-training with disagreement. In the experiments, we used 4 different text datasets for different tasks. We examined the algorithms from a variety of perspectives by asking experiment questions and suggested several improvements. Among the algorithms, tri-training with disagreement showed the closest performance to the Oracle; however, performance gap shows that new semi-supervised algorithms or improvements in existing methods are needed.