Prominent Features of Rumor Propagation in Online Social Media

By Sejeong Kwon, Meeyoung Cha, Kyomin Jung, Yajun Wang. 2013 IEEE 13th International Conference on Data Mining, 2013

796 citations recorded by the source corpus.

The problem of identifying rumors is of practical importance especially in online social networks, since information can diffuse more rapidly and widely than the offline counterpart. In this paper, we identify characteristics of rumors by examining the following three aspects of diffusion: temporal, structural, and linguistic. For the temporal characteristics, we propose a new periodic time series model that considers daily and external shock cycles, where the model demonstrates that rumor likely have fluctuations over time. We also identify key structural and linguistic differences in the spread of rumors and non-rumors. Our selected features classify rumors with high precision and recall in the range of 87% to 92%, that is higher than other states of the arts on rumor classification.

Subfields: Sociology and Political Science, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/e1682c302ac49647d50b97e6b2ead2fd87225fb3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Sociology and Political Science

In Statistical and Nonlinear Physics

Citations in the corpus

Detect Rumors Using Time Series of Social Context Information on Microblogging Websites (2015, methodology): “[5, 4] recently introduced a time series fitting model that shows better detection result based on the temporal properties regarding a single feature – tweet volume.”
Disinformation on the Web: Impact, Characteristics, and Detection of Wikipedia Hoaxes (2016, background): “[21] identify temporal, structural, and linguistic features of rumors on Twitter; Gupta et al.”
Fake News: A Survey of Research, Detection Methods, and Opportunities (2018, background): “…with image features [Jin et al. 2017a,b; Wang et al. 2018], user features [Jin et al. 2017a; Tacchini et al. 2017], cascade/network features [Kwon et al. 2013; Wu et al. 2015], temporal features [Kwon et al. 2013; Zhou et al. 2015], among other features to form a comprehensive set of…”
Fake News Detection on Social Media: A Data Mining Perspective (2017, methodology): “g, number of posts), mathematical features can be computed, such as SpikeM parameters [42].”
Multimodal Fusion with Recurrent Neural Networks for Rumor Detection on Microblogs (2017, background): “Semantic text features represent abstract semantics of texts, which include sentiment scores [7] and opinion words [19].”
A Survey of Fake News (2020, methodology): “…a news cascade that has been represented as a set of features, one often relies on supervised learning methods such as SVMs [Castillo et al. 2011; Kwon et al. 2013; Wu et al. 2015], decision trees [Castillo et al. 2011; Kwon et al. 2013], decision rules [Castillo et al. 2011], naive Bayes…”
Novel Visual and Statistical Image Features for Microblogs News Verification (2017, methodology): “While most existing studies present the evaluation data set with ground truth defined by human annotators [3], [5], we use the decision of authoritative sources to form a convincing ground truth of the truth value of news events in the data set.”
News Verification by Exploiting Conflicting Social Viewpoints in Microblogs (2016, methodology): “Kwon(2013) (Kwon et al. 2013; Wu, Yang, and Zhu 2015; Gupta et al. 2013): A classiﬁcation method at tweet level.”
EANN: Event Adversarial Neural Networks for Multi-Modal Fake News Detection (2018, background): “Textual features are statistical or semantic features extracted from text content of posts, which have been explored in many literatures of fake news detection [4, 11, 19, 27].”
Rumor Detection on Social Media with Bi-Directional Graph Convolutional Networks (2020, methodology): “…detection methods mainly adopt hand-crafted features such as user characteristics, text contents and propagation patterns to train supervised classiﬁers, e.g., Decision Tree (Castillo, Mendoza, and Poblete 2011), Random Forest (Kwon et al. 2013), Support Vector Machine (SVM) (Yang et al. 2012).”
Rumor Detection on Twitter with Tree-structured Recursive Neural Networks (2018, methodology): “…studies focused on text mining from sequential microblog streams using supervised models based on feature engineering (Castillo et al., 2011; Kwon et al., 2013; Liu et al., 2015; Ma et al., 2015), and more recently deep neural models (Ma et al., 2016; Chen et al., 2017; Ruchan-sky et al.,…”
Detect Rumors in Microblog Posts Using Propagation Structure via Kernel Learning (2017, background): “Existing studies considering propagation characteristics mainly focused on the temporal features (Kwon et al., 2013, 2017) rather than the structure of propagation.”
Hierarchical Multi-modal Contextual Attention Network for Fake News Detection (2021, methodology): “Basically, existing studies on automatic fake news detection can be summarized into two categories: (1) The first one is traditional learning methods [2, 16, 18, 20], which design plenty of hand-crafted features from the media content of posts and the Session 1D: Social Aspects SIGIR ’21, July 11–15”
Fake News Identification on Twitter with Hybrid CNN and RNN Models (2018, background): “. 3.3 Recurrent Neural Network (RNNs) This type of neural network has been shown to be effective in time and sequence based predictions [25]. Twitter posts can be likened to events that occur in time [16] where the intervals between the retweet of one user to another is contained within a time windo”
Fake News Detection on Social Media using Geometric Deep Learning (2019, background): “So far, attempts to exploit news propagation for fake news detection applied ‘handcrafted’ graph-theoretical features such as centrality, cliques, or connected components [15].”