CED: Credible Early Detection of Social Media Rumors

Song, Changhe; Yang, Cheng; Chen, Huimin; Tu, Cunchao; Liu, Zhiyuan; Sun, Maosong

doi:10.1109/tkde.2019.2961675

Cited by 106 publications

(60 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With regards to neural networks, RNNs are used by Ma et al [30] to learn hidden representations of posts, without the need of extracting hand-crafted features. The task of early detection of social media rumours is investigated by Song et al [38] proposing a model called Credible Early Detection. In contrast to existing methods, which typically need all reposts of a rumour for making the prediction, this work aims to make credible predictions soon after the initial suspicious post.…”

Section: Machine Learning Approachesmentioning

confidence: 99%

Verification of Web Videos Through Analysis of Their Online Context

Papadopoulou

Zampoglou

Papadopoulos

et al. 2019

Video Verification in the Fake News Era

View full text Add to dashboard Cite

This chapter discusses the problem of analysing the online "context" of User Generated Videos (UGVs) with the goal of extracting clues that help analysts with the video verification process. As video context, we refer to information surrounding the video i.e. information about the video itself, user comments below the video, information about the video publisher and any dissemination of the video through other video platforms or social media. As a starting point, we present the Fake Video Corpus, a dataset of debunked and verified UGVs that aims at serving as reference for qualitative and quantitative analysis and evaluation. Next, we present a web-based service, called Context Aggregation and Analysis, which supports the collection, filtering and mining of contextual pieces of information that can serve as verification signals. This service aims to assist Internet users in their video verification efforts.

show abstract

Section: Machine Learning Approachesmentioning

confidence: 99%

Verification of Web Videos Through Analysis of Their Online Context

Papadopoulou

Zampoglou

Papadopoulos

et al. 2019

Video Verification in the Fake News Era

View full text Add to dashboard Cite

show abstract

“…In many of the rumour detection strategies it is found that machine learning approaches requires frequent training of system using the labelled dataset in case of supervised strategy. Studies suggest that, instead of machine learning or ranking algorithmic approach, the pre-defined knowledge based rules sum-up with deep learning approaches (CNN/RNN) and Natural Language processing embedded with Semantic Ontology are found to be efficient when compared to earlier approaches [16] [8]. After keen observation of various datasets of different domain it is suggested that domain specific rumour words need to be used in a particular context for better precision rate.…”

Section: Problems Faced Due To Rumours In Snsmentioning

confidence: 99%

“…Detecting of short rumour tweets such as -Obama is a muslim guy‖, this does not require excessive learning for this authenticated books to be checked and a conclusion can be derived within seconds using a pre-defined rule based classifier [13]. In literature, most of the researchers had picked datasets of their own considering a particular scenario and few have taken from previously existed rumour datasets [3] [13] [16]. Many of the researchers had picked rumour dataset of microblogs from Twitter, Facebook, Weibo and Instagram to predict the category of rumours [1] [18].…”

Section: A Rumour Datasetsmentioning

confidence: 99%

“…In this chatting session the rumour words like crash, disaster, tragedy, collision, survive and deathblow are used which needs to be detected. The above chatting session is tested with Rumour Detection strategies TraceMiner [15], CED [16] and the proposed RDM. The precision rate that are obtained are shown in Table IV.…”

Section: A Rumour Scenario In Social Networking Sitesmentioning

confidence: 99%

See 1 more Smart Citation

Rumour Detection Models & Tools for Social Networking Sites

Ali,

Qaseem,

Rahman

2019

IJEAT

View full text Add to dashboard Cite

Efficient utilization of social networking sites (SNS) had reduced communication delays, at the same time increased rumour messages. Subsequently, mischievous people started sharing of rumours via social networking sites for gaining personal benefits. This falsified information (i.e., rumour) creates misconception among the people of society influencing socio-economic losses by disrupting the routine businesses of private and government sectors. Communication of rumour information requires rigorous surveillance, before they become viral through social media platforms. Detecting these rumour words in an early stage from messaging applications needs to be predicted using robust Rumour Detection Models (RDM) and succinct tools. RDM are effectively used in detecting the rumours from social media platforms (Twitter, Linkedln, Instagram, WhatsApp, Weibo sena and others) with the help of bag of words and machine learning approaches to a limited extent. RDM fails in detecting the emerging rumours that contains linguistic words of a specific language during the chatting session. This survey compares the various RDM strategies and Tools that were proposed earlier for identifying the rumour words in social media platforms. It is found that many of earlier RDM make use of Deep learning approaches, Machine learning, Artificial Intelligence, Fuzzy logic technique, Graph theory and Data mining techniques. Finally, an improved RDM model is proposed in Figure 2, efficiency of this proposed RDM models is improved by embedding of Pre-defined rumour rules, WordNet Ontology and NLP/machine learning approach giving the precision rate of 83.33% when compared with other state-of-art systems.

show abstract

“…Motivated by the success of deep learning, many recent studies [8], [9] apply various neural networks for rumor detec-* is the corresponding author. tion.…”

Section: Introductionmentioning

confidence: 99%

Jointly Embedding the Local and Global Relations of Heterogeneous Graph for Rumor Detection

Yuan

Zhou

et al. 2019

2019 IEEE International Conference on Data Mining (ICDM)

137

View full text Add to dashboard Cite

The development of social media has revolutionized the way people communicate, share information and make decisions, but it also provides an ideal platform for publishing and spreading rumors. Existing rumor detection methods focus on finding clues from text content, user profiles, and propagation patterns. However, the local semantic relation and global structural information in the message propagation graph have not been well utilized by previous works.In this paper, we present a novel global-local attention network (GLAN) for rumor detection, which jointly encodes the local semantic and global structural information. We first generate a better integrated representation for each source tweet by fusing the semantic information of related retweets with the attention mechanism. Then, we model the global relationships among all source tweets, retweets, and users as a heterogeneous graph to capture the rich structural information for rumor detection. We conduct experiments on three real-world datasets, and the results demonstrate that GLAN significantly outperforms the state-ofthe-art models in both rumor detection and early detection scenarios.

show abstract

CED: Credible Early Detection of Social Media Rumors

Cited by 106 publications

References 27 publications

Verification of Web Videos Through Analysis of Their Online Context

Verification of Web Videos Through Analysis of Their Online Context

Rumour Detection Models & Tools for Social Networking Sites

Jointly Embedding the Local and Global Relations of Heterogeneous Graph for Rumor Detection

Contact Info

Product

Resources

About