Rethinking Negative Sampling for Handling Missing Entity Annotations
This work addresses missing annotations in NER, an incremental improvement for researchers and practitioners in natural language processing.
The paper tackles the problem of missing entity annotations in named entity recognition by analyzing negative sampling, showing that low missampling rate and high uncertainty are key for performance, and proposes an adaptive weighted sampling distribution that improves F1 scores and achieves state-of-the-art results on datasets like EC.
Negative sampling is highly effective in handling missing annotations for named entity recognition (NER). One of our contributions is an analysis on how it makes sense through introducing two insightful concepts: missampling and uncertainty. Empirical studies show low missampling rate and high uncertainty are both essential for achieving promising performances with negative sampling. Based on the sparsity of named entities, we also theoretically derive a lower bound for the probability of zero missampling rate, which is only relevant to sentence length. The other contribution is an adaptive and weighted sampling distribution that further improves negative sampling via our former analysis. Experiments on synthetic datasets and well-annotated datasets (e.g., CoNLL-2003) show that our proposed approach benefits negative sampling in terms of F1 score and loss convergence. Besides, models with improved negative sampling have achieved new state-of-the-art results on real-world datasets (e.g., EC).