Residual convolutional graph neural network with subgraph attention pooling

Duan, Yutai; Wang, Jianming; Ma, Haoran; Sun, Yukuan

doi:10.26599/tst.2021.9010058

Cited by 19 publications

(7 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Literature [3] also proposed a GNN model that can capture the temporal characteristics of stocks using a variety of different types of relations in public knowledge databases. Literature [4] established a company network based on financial investment information.…”

Section: Introductionmentioning

confidence: 99%

Multilayer Graph-Based Deep Learning Approach for Stock Price Prediction

Alshara

2022

Security and Communication Networks

View full text Add to dashboard Cite

Although a strategy for predicting stock prices using relational data has been described recently, no practical way for selecting aggregating various forms of relational data to forecast stock prices has been discovered. The authors present an upgraded multilayer node graph attention network (FHAN) model that incorporates the Fraudar algorithm and provides insight into the interaction between several items. The model, which regards businesses as nodes and interactions as edges, aggregates data from various connection types and adds it to each company’s node representation, which is then automatically fed into the task-specific layer select information. The testing findings indicate that this approach is more accurate in stock price prediction than the currently popular neural network methodology. The experiment compares several distinct neural network algorithms. This approach is more accurate than the previous method when ideal parameters are used. The average rise is around 4%, while the largest increase is 24%.

show abstract

Section: Introductionmentioning

confidence: 99%

Multilayer Graph-Based Deep Learning Approach for Stock Price Prediction

Alshara

2022

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…Due to its good robustness, it is widely used in various practical scenarios. Among the various variant structures of GNN, Graph Convolutional Networks (GCNs) are the most widely used [13][14]. This model combines graph and pool, and uses graph convolution technology to analyze and process Big data such as images and videos, so as to obtain the semantic content hidden in them.…”

Section: Visual Feature Learningmentioning

confidence: 99%

New Media Detection Technology Based on Big Data Network Intelligent Algorithm

Xiao

2024

Proceedings of the First International Conference on Science, Engineering and Technology Practices for Sustainable Development,

View full text Add to dashboard Cite

With the continuous development of new media information on the Internet, it has become a vital task to monitor the information spread in new media by using emerging artificial intelligence technologies, such as Big data and machine learning, in order to maintain network order. However, currently existing information feature detection models are mostly based on the characteristics of specific content itself. The challenges faced by this type of information feature detection model include high risk of overfitting, insufficient generalization ability, and failure to fully explore the deep correlation features inherent in media information. Based on this, this article first uses the BERT model to extract multi-level semantics from social media post text content, and then uses GNN technology to obtain the visual characteristics of the image. It can adopt the GCN algorithm for optimization to improve the performance of GNN, which is difficult to quickly converge when facing complex social network topologies. Finally, this paper constructs an illegal information detection model based on Big data network intelligent algorithm. Experiments show that, compared with the two baseline methods of SVM-TS (Support Vector Machine Model for Tabu Search) and EANN (Event Adverse Neural Networks), the illegal information intelligent detection model proposed in this paper shows higher accuracy in identifying false information, rumors, real events and unconfirmed events.

show abstract

“…GCNs generalize the traditional convolution to graph‐structured data and have achieved extraordinary performance. They are widely used in many domains such as node embedding, 35 graph classification, 36 and link prediction 37 . GCNs fall into two main categories, that is, spectral and spatial approaches 38 .…”

Section: Related Workmentioning

confidence: 99%

STGNet: Short‐term residential load forecasting with spatial–temporal gated fusion network

Feng,

Li,

Zhou

et al. 2023

Energy Science & Engineering

View full text Add to dashboard Cite

With the continuous transformation of the global power system, ensuring the flexibility and stability of the power system has become a direction that the global energy industry has been striving for. This puts forward higher requirements for load forecasting, especially high‐precision short‐term residential load forecasting. However, the high volatility and uncertainty of electric consumption make this problem challenging. Existing methods usually focus on capturing temporal correlations and ignore the underlying spatial correlations in historical loads, which is crucial for accurate forecasting. In this paper, we propose a novel spatial–temporal gated fusion network (STGNet) for short‐term residential load forecasting including not only individual residential load but also aggregated load. By seeking both adaptive neighbors and temporal similarity neighbors, we construct a dynamic graph based on a data‐driven method. On this basis, we provide an adaptive graph gated fusion mechanism to capture fine‐grained node‐specific spatial dependence more accurately and comprehensively. Finally, a bidirectional gated recurrent unit (BiGRU) is utilized to learn the temporal dependence of load data. Extensive experiments on real‐world datasets are conducted and demonstrate the superiority of STGNet over the existing approaches.

show abstract

Residual convolutional graph neural network with subgraph attention pooling

Cited by 19 publications

References 8 publications

Multilayer Graph-Based Deep Learning Approach for Stock Price Prediction

Multilayer Graph-Based Deep Learning Approach for Stock Price Prediction

New Media Detection Technology Based on Big Data Network Intelligent Algorithm

STGNet: Short‐term residential load forecasting with spatial–temporal gated fusion network

Contact Info

Product

Resources

About