Exploring Segment Representations for Neural Segmentation Models
This addresses segmentation problems in NLP, such as named entity recognition and Chinese word segmentation, with incremental improvements over existing methods.
The paper tackles NLP segmentation tasks by combining semi-CRF with neural networks to represent segments through composition functions and embeddings, achieving state-of-the-art performance on Chinese word segmentation and competitive results on named entity recognition benchmarks.
Many natural language processing (NLP) tasks can be generalized into segmentation problem. In this paper, we combine semi-CRF with neural network to solve NLP segmentation tasks. Our model represents a segment both by composing the input units and embedding the entire segment. We thoroughly study different composition functions and different segment embeddings. We conduct extensive experiments on two typical segmentation tasks: named entity recognition (NER) and Chinese word segmentation (CWS). Experimental results show that our neural semi-CRF model benefits from representing the entire segment and achieves the state-of-the-art performance on CWS benchmark dataset and competitive results on the CoNLL03 dataset.