Rumor Detection with Bidirectional Graph Attention Networks

Yang, Xiaohui; Ma, Hailong; Wang, Miao

doi:10.1155/2022/4840997

Cited by 7 publications

(7 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… Bi-GCN [ 20 ]: A graph convolution neural network model based on the propagation direction and diffusion direction of the propagation tree. P-BiGAT [ 22 ]: A bidirectional graph attention networks based on the propagation tree and diffusion tree through the tweet comment and reposting relationship. …”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Rumor detection driven by graph attention capsule network on dynamic propagation structures

et al. 2022

View full text Add to dashboard Cite

Rumor detection aims to judge the authenticity of posts on social media (such as Weibo and Twitter), which can effectively prevent the spread of rumors. While many recent rumor detection methods based on graph neural networks can be conducive to extracting the global features of rumors, each node of the rumor propagation structure learned from graph neural networks is considered to have multiple individual scalar features, which are insufficient for reflecting the deep-level rumor properties. To address the above challenge, we propose a novel model named graph attention capsule network on dynamic propagation structures (GACN) for rumor detection. Specifically, GACN consists of two components: a graph attention network enforced by capsule network that can encode static graphs into substructure classification capsules for mining the deep-level properties of rumor, and a dynamic network framework that can divide the rumor structure into multiple static graphs in chronological order for capturing the dynamic interactive features in the evolving process of the rumor propagation structure. Moreover, we use the capsule attention mechanism to combine the capsules generated from each substructure to focus more on informative substructures in rumor propagation. Extensive validation on two real-world datasets demonstrates the superiority of GACN over baselines.

show abstract

Section: Methodsmentioning

confidence: 99%

“…P-BiGAT [ 22 ]: A bidirectional graph attention networks based on the propagation tree and diffusion tree through the tweet comment and reposting relationship.…”

Section: Methodsmentioning

confidence: 99%

Rumor detection driven by graph attention capsule network on dynamic propagation structures

et al. 2022

View full text Add to dashboard Cite

show abstract

“…K k�1 represented the concatenation of vectors from x 1 to x k , and W l k was a trainable weight matrix. e a ij represented the attention coefficient of the node j , use the same method as literature [24] calculated as in the following formula:…”

Section: Graph Attention Network(gat)mentioning

confidence: 99%

Rumor Detection Based on Knowledge Enhancement and Graph Attention Network

Wang

Wen

et al. 2022

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

Presently, most of the existing rumor detection methods focus on learning and integrating various features for detection, but due to the complexity of the language, these models often rarely consider the relationship between the parts of speech. For the first time, this paper integrated a knowledge graphs and graph attention networks to solve this problem through attention mechanisms. A knowledge graphs can be the most effective and intuitive expression of relationships between entities, providing problem analysis from the perspective of “relationships”. This paper used knowledge graphs to enhance topics and learn the text features by using self-attention. Furthermore, this paper defined a common dependent tree structure, and then the ordinary dependency trees were reshaped to make it generate a motif-dependent tree. A graph attention network was adopted to collect feature representations derived from the corresponding syntax-dependent tree production. The attention mechanism was an allocation mechanism of weight parameters that could help the model capture important information. Rumors were then detected accordingly by using the attention mechanism to combine text representations learned from self-attention and graph representations learned from the graph attention network. Finally, numerous experiments were performed on the standard dataset Twitter, and the proposed model here had achieved a 7.7% improved accuracy rate compared with the benchmark model.

show abstract

“…MapReduce [13][14][15] parallel computing framework is a parallel computing model running on HDFS distributed storage system [16]. It can process large PB-level data in parallel in a high fault-tolerant way and realize the parallel task processing function of Hadoop platform [17,18]. The core idea of this design is to divide and conquer the problem, not to push the data to the calculation, but to push the calculation to the data, which can greatly reduce the communication overhead.…”

Section: Improvement Of Dual Threshold Single-pass Algorithmmentioning

confidence: 99%

Internet Rumor Audience Response Prediction Algorithm Based on Machine Learning in Big Data Environment

Yang

Wang

Yiwen

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Rumors are an important factor affecting social stability in some special times. Therefore, the dissemination and prevention and control mechanisms of rumors have always been issues of concern to the academic community and have long been highly valued and widely discussed by experts and scholars. However, in combination with the Internet as a new type of media, although people have begun to pay attention to online rumors, research on it is still relatively fragmented, especially in the cross-domain research specific to the social influence of online rumors, and there is no clear indication of online rumors. The specific definition also did not analyze in detail the internal connection between its influence and group behavior. Therefore, this article will combine actual cases to explore and analyze the spread and influence process of online rumors and show its social influence, hoping to enrich the research of online rumors. Nowadays, the Internet has become the most important carrier to reflect the public grievances. Internet users have expressed their opinions on hot issues such as enterprises, people’s livelihood, and government management, which has formed a powerful public opinion pressure, which has far exceeded the traditional media. The hidden dangers of security cannot be ignored. Therefore, how to monitor network public opinion from a large amount of network data is a difficult problem that needs to be solved urgently. Firstly, this consists of four modules: information collection, web page preprocessing, public opinion analysis, and public information report. Secondly, text clustering, the core technology of network public opinion, is optimized, and single-pass algorithm based on double threshold is proposed. Then the dual-threshold single-pass algorithm is optimized based on the MapReduce parallel computing model, and finally a network public opinion collection technology is formed under the background of big data. Simulation results can greatly improve the performance of text clustering and can effectively optimize the design using the parallel computing model based on MapReduce. The average miss rate after optimization is 0.7569 times, the average false alarm rate is 0.5556 times, and C det is 0.5714 times. It proves that the collection technology based on machine learning under the background of big data is effective and has good performance.

show abstract

Rumor Detection with Bidirectional Graph Attention Networks

Cited by 7 publications

References 33 publications

Rumor detection driven by graph attention capsule network on dynamic propagation structures

Rumor detection driven by graph attention capsule network on dynamic propagation structures

Rumor Detection Based on Knowledge Enhancement and Graph Attention Network

Internet Rumor Audience Response Prediction Algorithm Based on Machine Learning in Big Data Environment

Contact Info

Product

Resources

About