Automatically extracting causal relations from texts is a challenging task in Natural Language Processing (NLP). Most existing methods focus on extracting intra-sentence or explicit causality, while neglecting the causal relations that expressed implicitly or hidden in inter-sentences. In this paper, we propose Cascaded multi-Structure Neural Network (CSNN), a novel and unified model that extract intersentence or implicit causal relations from Chinese Corpus, without relying on external knowledge. The model employs Convolutional Neural Network (CNN) to capture important features as well as causal structural pattern. Self-attention mechanism is designed to mine semantic and relevant characteristics between different features. The output of CNN and self-attention structure are concatenated as higher-level phrase representations. Then Conditional Random Field (CRF) layer is employed to calculate the label of each word in inter-sentence or implicit causal relation sentences, which improves the performance of inter-sentence or implicit causality extraction. Experimental results show that the proposed model achieves state-of-the-art results, improved on three datasets, when compared with other methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.