No abstract

No abstract

Background: Most Chinese joint entity and relation extraction tasks in medicine involve numerous nested entities, overlapping relations, and other challenging extraction issues. In response to these problems, some traditional methods decompose the joint extraction task into multiple steps or multiple modules, resulting in local dependency in the meantime. Methods: To alleviate this issue, we propose a joint extraction model of Chinese medical entities and relations based on RoBERTa and single-module global pointer, namely RSGP, which formulates joint extraction as a global pointer linking problem. Considering the uniqueness of Chinese language structure, we introduce the RoBERTa-wwm pre-training language model at the encoding layer to obtain a better embedding representation. Then, we represent the input sentence as a three-dimensional matrix and score each position in the matrix to prepare for the subsequent process of decoding the triples. In the end, we design a novel single-module global pointer decoding approach to alleviate the generation of redundant information. Specifically, we analyze the decoding process of single character entities individually, improving the time and space performance of RSGP to some extent. Results: In order to verify the effectiveness of our model in extracting Chinese medical entities and relations, we carry out the experiments on the public dataset, CMeIE. Experimental results show that RSGP performs significantly better on the joint extraction of Chinese medical entities and relations, and achieves state-of-the-art results compared with baseline models. Conclusion: The proposed RSGP can effectively extract entities and relations from Chinese medical texts and help to realize the structure of Chinese medical texts, so as to provide high-quality data support for the construction of Chinese medical knowledge graphs.

Text classification is a research hotspot in the field of natural language processing. Existing text classification models based on supervised learning, especially deep learning models, have made great progress on public datasets. But most of these methods rely on a large amount of training data, and these datasets coverage is limited. In the legal intelligent question-answering system, accurate classification of legal consulting questions is a necessary prerequisite for the realization of intelligent question answering. However, due to lack of sufficient annotation data and the cost of labeling is high, which lead to the poor effect of traditional supervised learning methods under sparse labeling. In response to the above problems, we construct a few-shot legal consulting questions dataset, and propose a prototypical networks model based on multi-attention. For the same category of instances, this model first highlights the key features in the instances as much as possible through instance-dimension level attention. Then it realizes the classification of legal consulting questions by prototypical networks. Experimental results show that our model achieves state-of-the-art results compared with baseline models. The code and dataset are released on https://github.com/cjm0824/MAPN.

Pre-trained language models achieve high performance on machine reading comprehension task, but these models lack robustness and are vulnerable to adversarial samples. Most of the current methods for improving model robustness are based on data enrichment. However, these methods do not solve the problem of poor context representation of the machine reading comprehension model. We find that context representation plays a key role in the robustness of the machine reading comprehension model, dense context representation space results in poor model robustness. To deal with this, we propose a Multi-task machine Reading Comprehension learning framework via Contrastive Learning. Its main idea is to improve the context representation space encoded by the machine reading comprehension models through contrastive learning. This special contrastive learning we proposed called Contrastive Learning in Context Representation Space(CLCRS). CLCRS samples sentences containing context information from the context as positive and negative samples, expanding the distance between the answer sentence and other sentences in the context. Therefore, the context representation space of the machine reading comprehension model has been expanded. The model can better distinguish between sentence containing correct answers and misleading sentence. Thus, the robustness of the model is improved. Experiment results on adversarial datasets show that our method exceeds the comparison models and achieves state-of-the-art performance.