Early Rumor Detection Based on Deep Recurrent Q-Learning

Abstract

Online social networks provide convenient conditions for the spread of rumors, and false rumors bring great harm to social life. Rumor dissemination is a process, and effective identification of rumors in the early stage of their appearance will reduce the negative impact of false rumors. This paper proposes a novel early rumor detection (ERD) model based on reinforcement learning. In the rumor detection part, a dual-engine rumor detection model based on deep learning is proposed to realize the differential feature extraction of original tweets and their replies. A double self-attention (DSA) mechanism is proposed, which can eliminate data redundancy in sentences and words at the same time. In the reinforcement learning part, an ERD model based on Deep Recurrent Q-Learning Network (DRQN) is proposed, which uses LSTM to learn the state sequence features, and the optimization strategy of the reward function is to take into account the timeliness and accuracy of rumor detection. Experiments show that, compared with existing methods, the ERD model proposed in this paper has a greater improvement in the timeliness and detection rate of rumor detection.

A Rumor Detection Method from Social Network Based on Deep Learning in Big Data Environment

“…Moreover, there are a large number of unidentified and even contradictory information on the network, which also makes users more prone to anxiety and confusion. With the help of online social media, rumors will spread rapidly [ 15 – 18 ]. These spreading rumors will greatly mislead the people become a potential unstable factor in society and affect the stable operation of economy and society.…”

A Hybrid Approach for Network Rumor Detection Based on Attention Mechanism and Bidirectional GRU Model in Big Data Environment

“…[17], Ref. [18], and the proposed model) differ when the learning rate is different, and the results of different evaluation index are shown in Figure 3.…”

“…[17] and Ref. [18] will be degraded due to the possible overfitting state in training. Therefore, in the parameter setting, the experimental demonstration is carried out when the selected learning rate is 0.015.…”

“…[17], and Ref. [18] based on the selected microblog dataset, and the values of each evaluation index are shown in Table 2.…”

“…A novel early rumor detection model based on reinforcement learning was proposed in Ref. [18]. The dual-engine rumor detection model based on deep learning can achieve differential feature extraction of the original tweets and their replies.…”