FRD: Few-shot rumor detection for the novel coronavirus pneumonia epidemic

With the full arrival of the digital era, fueled by both information technology and business marketing, rumors are produced and spread endlessly on social networks. During the recent novel coronavirus pneumonia epidemic, online rumors have continued to flourish. Most existing studies on traditional rumor detection rely on a large number of features in practical applications. However, the current severe epidemic scenarios have limited rumor information features, and it remains a challenging problem to detect epidemic rumors with high accuracy using only limited information. As a result, we propose a novel Few-shot Rumor Detection model (FRD) for the novel coronavirus pneumonia, which is combined with meta-learning to be able to accurately identify rumors as soon as possible in crises. Specifically, we started by using the BERT+BiLSTM combination for rumor text feature extraction and representation to generate the historical rumor sample-wise vector and epidemic rumor sample-wise vector;secondly, the prototypical network was introduced to summarize the historical rumor data, and the feature vectors of samples belonging to the same category were averaged to obtain the prototype representation of historical rumor category;finally, we utilize the modified cosine similarity measure function to calculate the distance between the class-wise vector of historical rumor text and the sample-wise vector of epidemic rumor, and complete the rumor detection according to the nearest neighbor method. Our experimental results on English datasets show that the FRD rumor detection model proposed in this paper is superior to other baseline algorithms in terms of accuracy, precision, recall and macro F1 value. From the comparison of experimental results, the FRD model can effectively improve conventional rumor detection methods, and better realize the early detection of sudden epidemic rumors. © 2022 IEEE.

Rumor Localization, Detection and Prediction in Social Network

With the global epidemic of the COVID-19, various rumors spread wantonly on social networks, which has seriously affected the stability and harmony of the entire society. To purify the network environment, some researchers have proposed to fight rumors from the perspectives of tracing the source of rumors, detecting the authenticity of information, and predicting explosive fake news. But their works are fragmented, and their performance are not significant. So we need strong antirumor methods to fight rumors. To this end, this article proposes a more comprehensive antirumor mechanism, which can realize rumors source location, rumor detection, and popularity prediction (RLDP). In particular, in the task of localization, we propose graph neural network-based method, which does not need to specify the underlying propagation mode and the number of rumor sources;in the task of detection, utilizing lightGBM, we construct a rumor detection model;in the task of popularity prediction, we construct a model based on contrastive learning while considering user engagements and information propagation, and the text of rumor. Finally, we verify the performance of the proposed RLDP by conducting extensive experiments. IEEE

A novel rumor detection with multi-objective loss functions in online social networks.

COVID-19 quickly swept across the world, causing the consequent infodemic represented by the rumors that have brought immeasurable losses to the world. It is imminent to achieve rumor detection as quickly and accurately as possible. However, the existing methods either focus on the accuracy of rumor detection or set a fixed threshold to attain early detection that unfortunately cannot adapt to various rumors. In this paper, we focus on textual rumors in online social networks and propose a novel rumor detection method. We treat the detection time, accuracy and stability as the three training objectives, and continuously adjust and optimize this objective instead of using a fixed value during the entire training process, thereby enhancing its adaptability and universality. To improve the efficiency, we design a sliding interval to intercept the required data rather than using the entire sequence data. To solve the problem of hyperparameter selection brought by integration of multiple optimization objectives, a convex optimization method is utilized to avoid the huge computational cost of enumerations. Extensive experimental results demonstrate the effectiveness of the proposed method. Compared with state-of-art counterparts in three different datasets, the recognition accuracy is increased by an average of 7%, and the stability is improved by an average of 50%.

HAT4RD: Hierarchical Adversarial Training for Rumor Detection in Social Media.

With the development of social media, social communication has changed. While this facilitates people's communication and access to information, it also provides an ideal platform for spreading rumors. In normal or critical situations, rumors can affect people's judgment and even endanger social security. However, natural language is high-dimensional and sparse, and the same rumor may be expressed in hundreds of ways on social media. As such, the robustness and generalization of the current rumor detection model are in question. We proposed a novel hierarchical adversarial training method for rumor detection (HAT4RD) on social media. Specifically, HAT4RD is based on gradient ascent by adding adversarial perturbations to the embedding layers of post-level and event-level modules to deceive the detector. At the same time, the detector uses stochastic gradient descent to minimize the adversarial risk to learn a more robust model. In this way, the post-level and event-level sample spaces are enhanced, and we verified the robustness of our model under a variety of adversarial attacks. Moreover, visual experiments indicate that the proposed model drifts into an area with a flat loss landscape, thereby, leading to better generalization. We evaluate our proposed method on three public rumor datasets from two commonly used social platforms (Twitter and Weibo). Our experimental results demonstrate that our model achieved better results compared with the state-of-the-art methods.

A deep neural networks-based fusion model for COVID-19 rumor detection from online social media

Purpose The COVID-19 has become a global pandemic, which has caused large number of deaths and huge economic losses. These losses are not only caused by the virus but also by the related rumors. Nowadays, online social media are quite popular, where billions of people express their opinions and propagate information. Rumors about COVID-19 posted on online social media usually spread rapidly;it is hard to analyze and detect rumors only by artificial processing. The purpose of this paper is to propose a novel model called the Topic-Comment-based Rumor Detection model (TopCom) to detect rumors as soon as possible. Design/methodology/approach The authors conducted COVID-19 rumor detection from Sina Weibo, one of the most widely used Chinese online social media. The authors constructed a dataset about COVID-19 from January 1 to June 30, 2020 with a web crawler, including both rumor and non-rumors. The rumor detection task is regarded as a binary classification problem. The proposed TopCom model exploits the topical memory networks to fuse latent topic information with original microblogs, which solves the sparsity problems brought by short-text microblogs. In addition, TopCom fuses comments with corresponding microblogs to further improve the performance. Findings Experimental results on a publicly available dataset and the proposed COVID dataset have shown superiority and efficiency compared with baselines. The authors further randomly selected microblogs posted from July 1-31, 2020 for the case study, which also shows the effectiveness and application prospects for detecting rumors about COVID-19 automatically. Originality/value The originality of TopCom lies in the fusion of latent topic information of original microblogs and corresponding comments with DNNs-based models for the COVID-19 rumor detection task, whose value is to help detect rumors automatically in a short time.

Dual Co-Attention-Based Multi-Feature Fusion Method for Rumor Detection

Social media has become more popular these days due to widely used instant messaging. Nevertheless, rumor propagation on social media has become an increasingly important issue. The purpose of this study is to investigate the impact of various features in social media on rumor detection, propose a dual co-attention-based multi-feature fusion method for rumor detection, and explore the detection capability of the proposed method in early rumor detection tasks. The proposed BERT-based Dual Co-attention Neural Network (BDCoNN) method for rumor detection, which uses BERT for word embedding. It simultaneously integrates features from three sources: publishing user profiles, source tweets, and comments. In the BDCoNN method, user discrete features and identity descriptors in user profiles are extracted using a one-dimensional convolutional neural network (CNN) and TextCNN, respectively. The bidirectional gate recurrent unit network (BiGRU) with a hierarchical attention mechanism is used to learn the hidden layer representation of tweet sequence and comment sequence. A dual collaborative attention mechanism is used to explore the correlation among publishing user profiles, tweet content, and comments. Then the feature vector is fed into classifier to identify the implicit differences between rumor spreaders and non-rumor spreaders. In this study, we conducted several experiments on the Weibo and CED datasets collected from microblog. The results show that the proposed method achieves the state-of-the-art performance compared with baseline methods, which is 5.2% and 5% higher than the dEFEND. The F1 value is increased by 4.4% and 4%, respectively. In addition, this paper conducts research on early rumor detection tasks, which verifies the proposed method detects rumors more quickly and accurately than competitors.

A novel few-shot learning based multi-modality fusion model for COVID-19 rumor detection from online social media.

BACKGROUND: Rumor detection is a popular research topic in natural language processing and data mining. Since the outbreak of COVID-19, related rumors have been widely posted and spread on online social media, which have seriously affected people's daily lives, national economy, social stability, etc. It is both theoretically and practically essential to detect and refute COVID-19 rumors fast and effectively. As COVID-19 was an emergent event that was outbreaking drastically, the related rumor instances were very scarce and distinct at its early stage. This makes the detection task a typical few-shot learning problem. However, traditional rumor detection techniques focused on detecting existed events with enough training instances, so that they fail to detect emergent events such as COVID-19. Therefore, developing a new few-shot rumor detection framework has become critical and emergent to prevent outbreaking rumors at early stages. METHODS: This article focuses on few-shot rumor detection, especially for detecting COVID-19 rumors from Sina Weibo with only a minimal number of labeled instances. We contribute a Sina Weibo COVID-19 rumor dataset for few-shot rumor detection and propose a few-shot learning-based multi-modality fusion model for few-shot rumor detection. A full microblog consists of the source post and corresponding comments, which are considered as two modalities and fused with the meta-learning methods. RESULTS: Experiments of few-shot rumor detection on the collected Weibo dataset and the PHEME public dataset have shown significant improvement and generality of the proposed model.