Recognizing named entities (NEs) is commonly treated as a classification problem, and a class tag for a word or an NE candidate in a sentence is predicted. In recent neural network developments, deep structures that map categorized features into continuous representations have been adopted. Using this approach, a dense space saturated with high-order abstract semantic information is unfolded, and the prediction is based on distributed feature representations. In this paper, the positions of NEs in a sentence are represented as continuous values. Then, a regression operation is introduced to regress the boundaries of NEs in a sentence. Based on boundary regression, we design a boundary regression model to support nested NE recognition. It is a multiobjective learning framework that simultaneously predicts the classification score of an NE candidate and refines its spatial location in a sentence. This model was evaluated on the ACE 2005 Chinese and English corpus and the GENIA corpus. State-of-the-art performance was experimentally demonstrated for nested NE recognition, which outperforms related works about 5% and 2% respectively. Our model has the advantage to resolve nested NEs and support boundary regression for locating NEs in a sentence. By sharing parameters for predicting and locating, this model enables more potent nonlinear function approximators to enhance model discriminability.

The task to extract relations tries to identify relationships between two named entities in a sentence. Because a sentence usually contains several named entities, capturing structural information of a sentence is important to support this task. Currently, graph neural networks are widely implemented to support relation extraction, in which dependency trees are employed to generate adjacent matrices for encoding structural information of a sentence. Because parsing a sentence is error-prone, it influences the performance of a graph neural network. On the other hand, a sentence is structuralized by several named entities, which precisely segment a sentence into several parts. Different features can be combined by prior knowledge and experience, which are effective to initialize a symmetric adjacent matrix for a graph neural network. Based on this phenomenon, we proposed a feature combination-based graph convolutional neural network model (FC-GCN). It has the advantages of encoding structural information of a sentence, considering prior knowledge, and avoiding errors caused by parsing. In the experiments, the results show significant improvement, which outperform existing state-of-the-art performances.