Summary of Preserving Empirical Probabilities in Bert For Small-sample Clinical Entity Recognition, by Abdul Rehman et al.
Preserving Empirical Probabilities in BERT for Small-sample Clinical Entity Recognition
by Abdul Rehman, Jian Jun Zhang, Xiaosong Yang
First submitted to arxiv on: 5 Sep 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper investigates the impact of unbalanced labels in Named Entity Recognition (NER) tasks, particularly with BERT-based pre-trained models. It examines how different loss calculation and propagation mechanisms affect performance on randomized datasets, focusing on token classification. The authors propose strategies to improve token classification for clinical entity recognition, a highly imbalanced task. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper looks at how unbalanced labels in Named Entity Recognition (NER) can lead to biased models that don’t work well for minority classes. They study BERT-based pre-trained models and see how different ways of calculating loss affect performance on random datasets. The goal is to make NER better for tasks like recognizing clinical entities. |
Keywords
» Artificial intelligence » Bert » Classification » Named entity recognition » Ner » Token
