RESUMEN

Medical Dialogue Information Extraction (MDIE) is a promising task for modern medical care systems, which greatly facilitates the development of many real-world applications such as electronic medical record generation, automatic disease diagnosis, etc. Recent methods have firstly achieved considerable performance in Chinese MDIE but still suffer from some inherent limitations, such as poor exploitation of the inter-dependencies in multiple utterances, weak discrimination of the hard samples. In this paper, we propose a contrastive multi-utterance inference (CMUI) method to address these issues. Specifically, we first use a type-aware encoder to provide an efficient encode mechanism toward different categories. Subsequently, we introduce a selective attention mechanism to explicitly capture the dependencies among utterances, which thus constructs a multi-utterance inference. Finally, a supervised contrastive learning approach is integrated into our framework to improve the recognition ability for the hard samples. Extensive experiments show that our model achieves state-of-the-art performance on a public benchmark Chinese-based dataset and delivers significant performance gain on MDIE as compared with baselines. Specifically, we outperform the state-of-the-art results in F1-score by 2.27%, 0.55% in Recall and 3.61% in Precision (The codes that support the findings of this study are openly available in CMUI at https://github.com/jc4357/CMUI.).

Asunto(s)

Aprendizaje Profundo , Almacenamiento y Recuperación de la Información , Benchmarking , China , Registros Electrónicos de Salud

RESUMEN

Knowledge graph completion is an important technology for supplementing knowledge graphs and improving data quality. However, the existing knowledge graph completion methods ignore the features of triple relations, and the introduced entity description texts are long and redundant. To address these problems, this study proposes a multi-task learning and improved TextRank for knowledge graph completion (MIT-KGC) model. The key contexts are first extracted from redundant entity descriptions using the improved TextRank algorithm. Then, a lite bidirectional encoder representations from transformers (ALBERT) is used as the text encoder to reduce the parameters of the model. Subsequently, the multi-task learning method is utilized to fine-tune the model by effectively integrating the entity and relation features. Based on the datasets of WN18RR, FB15k-237, and DBpedia50k, experiments were conducted with the proposed model and the results showed that, compared with traditional methods, the mean rank (MR), top 10 hit ratio (Hit@10), and top three hit ratio (Hit@3) were enhanced by 38, 1.3%, and 1.9%, respectively, on WN18RR. Additionally, the MR and Hit@10 were increased by 23 and 0.7%, respectively, on FB15k-237. The model also improved the Hit@3 and the top one hit ratio (Hit@1) by 3.1% and 1.5% on the dataset DBpedia50k, respectively, verifying the validity of the model.

RESUMEN

Automatic extraction of chemical-induced disease (CID) relation from unstructured text is of essential importance for disease treatment and drug development. In this task, some relational facts can only be inferred from the document rather than single sentence. Recently, researchers investigate graph-based approaches to extract relations across sentences. It iteratively combines the information from neighbor nodes to model the interactions in entity mentions that exist in different sentences. Despite their success, one severe limitation of the graph-based approaches is the over-smoothing problem, which decreases the model distinguishing ability. In this paper, we propose CID-GCN, an effective Graph Convolutional Networks (GCNs) with gating mechanism, for CID relation extraction. Specifically, we construct a heterogeneous graph which contains mention, sentence and entity nodes. Then, the graph convolution operation is employed to aggregate interactive information on the constructed graph. Particularly, we combine gating mechanism with the graph convolution operation to address the over-smoothing problem. The experimental results demonstrate that our approach significantly outperforms the baselines.