2020
DOI: 10.1007/978-3-030-41299-9_19
|View full text |Cite
|
Sign up to set email alerts
|

Dual-Attention Graph Convolutional Network

Abstract: Graph convolutional networks (GCNs) have shown the powerful ability in text structure representation and effectively facilitate the task of text classification. However, challenges still exist in adapting GCN on learning discriminative features from texts due to the main issue of graph variants incurred by the textual complexity and diversity. In this paper, we propose a dual-attention GCN to model the structural information of various texts as well as tackle the graph-invariant problem through embedding two t… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
8
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 11 publications
(8 citation statements)
references
References 20 publications
0
8
0
Order By: Relevance
“…2 the top-1 classification accuracy on both settings, comparing it with other existing approaches. As illustrated, CCP outperforms previous state-of-the-art methods on this dataset, including other graph-oriented architectures[31,17], despite being more general and not specifically designed to only address action recognition settings.…”
mentioning
confidence: 82%
See 1 more Smart Citation
“…2 the top-1 classification accuracy on both settings, comparing it with other existing approaches. As illustrated, CCP outperforms previous state-of-the-art methods on this dataset, including other graph-oriented architectures[31,17], despite being more general and not specifically designed to only address action recognition settings.…”
mentioning
confidence: 82%
“…Lie Group [28] 50.1 52.8 HBRNN-L [5] 59.1 64.0 P-LSTM [24] 62.9 70.3 ST-LSTM+TS [19] 69.2 77.7 TGCNN [31] 71.4 82.9 Temporal Conv [12] 74.3 83.1 Deep STGC K [17] 74.9 86.3 C-CNN + MTLN [11] 79.6 84.8 CCP (our) 80.1 86.8…”
Section: Methods Cross Subject Cross Viewmentioning
confidence: 99%
“…Generally speaking, words that frequently appear together can express a specific context and reflect a specific event, which helps us to detect fake news. We collect word co-occurrence information through the sliding window, and calculate the weight by a popular measure of word association, point-wise mutual information (PMI) [26]. E ru reflects the rumor-user and user-user interactions, and describes the structural relationship between rumor propagation and diffusion.…”
Section: Multi-information Heterogeneous Graph Constructionmentioning
confidence: 99%
“…Recently, lots of researchers have studies dual-attention graph neural networks [ 37 , 38 ] and developed a serious of application for general spatio-temporal network in different urban traffic scene [ 39 , 40 ].…”
Section: Related Workmentioning
confidence: 99%