Exploring deep neural networks for rumor detection

Asghar, Muhammad Zubair; Habib, Ammara; Habib, Anam; Khan, Adil; Ali, Rehman; Khattak, Asad Masood

doi:10.1007/s12652-019-01527-4

Cited by 103 publications

(59 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The rumor class has only 34% of data samples, whereas the non-rumor class has 66% data samples. The findings of our research are in line with similar works by Zubiaga et al (2016), Ma et al (2016), Yu et al (2017), Ajao et al (2018), and Asghar et al (2019). A comparative result of our models with earlier works implemented on the same Pheme dataset (Zubiaga et al 2016) is shown in Table 7.…”

Section: Discussionsupporting

confidence: 91%

“…The first result on the same dataset was reported by Zubiaga et al (2016) recall of 0.84, and F 1 -score of 0.83 was reported by Ajao et al (2018) with LSTM-CNN based hybrid model. Asghar et al (2019) reported precision, recall, and F 1 -score of 0.86 using bidirectional LSTM-CNN model. In line with this study, our proposed attention-based LSTM model using hybrid features achieves a precision, recall, and F 1 -score of 0.88 as shown in bold in Table 7 which is better than the existing state-of-the-art results.…”

Section: Discussionmentioning

confidence: 99%

“…The performance of these systems depended heavily on how efficiently the features were extracted. Recently, some deep learningbased models (Ma et al 2016;Chen et al 2017;Ajao et al 2018;Asghar et al 2019;Chen et al 2018a) have been proposed to reduce the limitations of handcrafted features to identify rumor messages. Ma et al (2016) used a recurrent neural network (RNN) model to predict the veracity of the social media post by automatic feature learning and semantic information learning capability.…”

Section: Literature Reviewmentioning

confidence: 99%

“…They found that good accuracy can be achieved with deep neural network-based models in case of rumor detection, even with a small amount of training data. Asghar et al (2019) proposed a deep neural network based on Bidirectional Long-Short Term Memory with Convolutional Neural Network (BiLSTM-CNN) to classify tweets into rumor and non-rumor classes. They have achieved a state-of-the-art result with the publicly available Pheme (Zubiaga et al 2016) dataset.…”

Section: Literature Reviewmentioning

confidence: 99%

“…On the other hand, deep learning models with automatic feature extracted from the text was also found to reach a limit as can be seen in Table 7. The hybrid deep learning models with a combination of LSTM (BiL-STM) and CNN by Ajao et al (2018) and Asghar et al (2019) achieved a precision of 0.83 and 0.86 respectively. Even our attention model also achieved the same result with a precision of 0.86.…”

Section: Theoretical Contributionsmentioning

confidence: 99%

See 4 more Smart Citations

Attention-Based LSTM Network for Rumor Veracity Estimation of Tweets

et al. 2020

View full text Add to dashboard Cite

Twitter has become a fertile place for rumors, as information can spread to a large number of people immediately. Rumors can mislead public opinion, weaken social order, decrease the legitimacy of government, and lead to a significant threat to social stability. Therefore, timely detection and debunking rumor are urgently needed. In this work, we proposed an Attention-based Long-Short Term Memory (LSTM) network that uses tweet text with thirteen different linguistic and user features to distinguish rumor and non-rumor tweets. The performance of the proposed Attention-based LSTM model is compared with several conventional machine and deep learning models. The proposed Attention-based LSTM model achieved an F 1-score of 0.88 in classifying rumor and non-rumor tweets, which is better than the state-of-the-art results. The proposed system can reduce the impact of rumors on society and weaken the loss of life, money, and build the firm trust of users with social media platforms.

show abstract

Section: Discussionsupporting

confidence: 91%

Section: Discussionmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Theoretical Contributionsmentioning

confidence: 99%

See 3 more Smart Citations

Attention-Based LSTM Network for Rumor Veracity Estimation of Tweets

et al. 2020

View full text Add to dashboard Cite

show abstract

Rumor, misinformation among web: A contemporary review of rumor detection techniques during different web waves

Rani

Das

Bhardwaj

2021

Concurrency and Computation

View full text Add to dashboard Cite

Summary Sometimes, unverified information is disseminated as if it is true information on social media sites. Most of the times, it goes viral and affects the belief of people and their emotions. Rumors and fake news are the most popular form of false and unconfirmed information. Such news must be identified quickly for preventing its negative impact on society. In the last decade, operational procedures for rumors and false news detection came into existence. This paper provides a holistic view of different web waves from web 1.0 to web 5.0 and their usages. Further, taxonomy describes various malicious information contents at different stages. It discusses features used for classification, publicly available datasets, the rumor detection methods proposed during web 1.0, 2.0, 3.0, 4.0, 5.0 periods, and comprehensive analysis related to various methods and techniques. Numerous research gaps and future directions are illustrated to make online information more trustworthy for knowledge sharing and decision‐making purposes.

show abstract

An effective veracity management for sentiment analysis in social media using crowdsourcing and multi‐set feature learning

Aruna¹,

Sangeetha²

2022

Concurrency and Computation

View full text Add to dashboard Cite

Recent innovations of the Internet-of-Things (IoT) are facilitated the data processing to an enormous scale. Nevertheless, the collected data are inherently uncertain because of noise, discrepancy, and incompleteness. To solve these constraints, improving data veracity is crucial for IoT-based applications. As big data veracity is more critical, it must predict the veracity of tweets automatically for decision-making. There have been many techniques developed including artificial intelligence and crowdsourcing models to estimate the veracity of tweets from single sources. But, these models are ineffective while the amount of crowd labels for tweets from multiple sources is not feasible. Also, analyzing the sentiment of public opinions has high complexity. Therefore, this article focuses on developing an efficient model to predict the veracity of tweets from multiple sources and analyze the sentiments of the passage. Primarily, different tweets with text labels are collected. Afterward, a crowdsourcing model is introduced which utilizes the crowdflower dataset to obtain the crowd labels for collected tweets. The Generative Adversarial Network (GAN) model is applied to increase the veracity of tweets by handling the low-veracity tweets. Finally, the multi-set feature learning model is implemented to classify the sentiment of opinions. The performance results manifest that the proposed model accomplishes superior accuracy of 97% than the classical veracity prediction models.

show abstract

Exploring deep neural networks for rumor detection

Cited by 103 publications

References 51 publications

Attention-Based LSTM Network for Rumor Veracity Estimation of Tweets

Attention-Based LSTM Network for Rumor Veracity Estimation of Tweets

Rumor, misinformation among web: A contemporary review of rumor detection techniques during different web waves

An effective veracity management for sentiment analysis in social media using crowdsourcing and multi‐set feature learning

Contact Info

Product

Resources

About