Deep Recurrent Neural Network and Data Filtering for Rumor Detection on Sina Weibo

Abstract

Social media makes it easy for individuals to publish and consume news, but it also facilitates the spread of rumors. This paper proposes a novel deep recurrent neural model with a symmetrical network architecture for automatic rumor detection in social media such as Sina Weibo, which shows better performance than the existing methods. In the data preparing phase, we filter the posts according to the followers of the user. We then use sequential encoding for the posts and multiple embedding layers to get better feature representation, and multiple recurrent neural network layers to capture the dynamic temporal signals characteristic. The experimental results on the Sina Weibo dataset show that: 1. the sequential encoding performs better than the term frequency-inverse document frequency (TF-IDF) or the doc2vec encoding scheme; 2. the model is more accurate when trained on the posts from the users with more followers; and 3. the model achieves superior improvements over the existing works on the accuracy of detection, including the early detection.

Deep learning for misinformation detection on online social networks: a survey and new perspectives

“…They argue that realworld data are structured sequences, with spatio-temporal sequences. For example, several works utilized a blend of CNN and RNN such as spatial and temporal regularities (Lin et al 2019;Xu et al 2019;Wang et al 2019). Their models can process time-shifting visual contributions for variable length expectations.…”

“…Figure 7 shows the structure of a deep hybrid model for fake news detection proposed by Ruchansky et al (2017). Xu et al (2019) proposed a CRNN model to extract data from textual overlays, for example, captions, key ideas, or scene level summaries for rumor detection on Sina Weibo. They proposed this CRNN model to create training data intended for textual overlays regularly occurring in the online sina weibo platform.…”

“…However, the automated detection of misinformation is difficult to accomplish as it requires the advanced model to understand how related or unrelated the reported information is when compared to real information (Wu et al 2019). Also, to solve many complex MID problems, academia and industry researchers have applied DL to a large number of applications to make decisions (Xu et al 2019;Yenala et al 2018;Yin et al 2020). Therefore, this survey seeks to provide such a systematic review of current research on MID based on DL techniques.…”

“…Also, to solve many complex MID problems, academia and industry researchers have applied DL to a large number of applications to make decisions (Xu et al. 2019 ; Yenala et al. 2018 ; Yin et al 2020 ).…”

Rumor Detection Based on Attention CNN and Time Series of Context Information

“…In recent years, deep neural networks have become more and more popular and they have achieved good performance in many NLP tasks. Y. Xu et al [20] used multiple RNNs to deeply mine dynamic time features and modeled the social background information of events as N equal time series, so as to obtain text sequence coding and capture the background information of relevant posts over time. On this basis, N. Ruchansky et al [21] divided the whole rumor recognition model into three modules: Capture, Score and Integrate, in which the Capture module uses RNN to learn the time representation of the text.…”