Social Media Event Prediction using DNN with Feedback Mechanism

Abstract

Online social networks (OSNs) are a rich source of information, and the data (including user-generated content) can be mined to facilitate real-world event prediction. However, the dynamic nature of OSNs and the fast-pace nature of social events or hot topics compound the challenge of event prediction. This is a key limitation in many existing approaches. For example, our evaluations of six baseline approaches (i.e., logistic regression latent Dirichlet allocation based logistic regression, multitask learning, long short-term memory and convolutional neural networks, and transformer-based model) on three datasets collected as part of this research (two from Twitter, and one from a news collection site 1 ), reveal that the accuracy of these approaches is between \(50\% \) and \(60\% \) , and they are not capable of utilizing new events in event predictions. Hence, in this paper we develop a novel DNN-based framework (hereafter referred to as event prediction with feedback mechanism – EPFM . Specifically, EPFM makes use of a feedback mechanism based on emerging events detection to improve the performance of event prediction. The feedback mechanism ensembles three outlier detection processes and returns a list of new events. Some of the events will then be chosen by analysts to feed into the fine-tuning process to update the predictive model. To evaluate EPFM, we conduct a series of experiments on the same three datasets, whose findings show that EPFM achieves \(80\% \) accuracy in event detection and outperforms the six baseline approaches.We also validate EPFM’s capability of detecting new events by empirically analyzing the feedback mechanism under different thresholds.

Design of New Media Event Warning Method Based on K-means and Seasonal Optimization Algorithm

“…Step 4: Compute the objective function using equation (1). Verify that the function converges or that the difference between two neighboring values of the objective function is less than a given threshold and stop.…”

“…In recent years, with the continuous development of network communication technology, the awakening of netizens' consciousness and the gradual opening of the democratic system, a series of new media events have appeared with the New Media (The New Media) as the carrier, and the power of netizens from all walks of life is widely involved in and disseminated, which has caused a significant social impact [1]. At present, the world is entering an era of frequent new media events, and China is in the stage of socio-economic transformation, the continuous development of new media events can easily lead to mass incidents [2].…”

Edge‐featured graph attention network with dependency features for causality detection of events

“…Event causality extraction, which is a subtask of information extraction (Claro et al, 2019), can accurately reflect the interaction between events and help people deepen their logical understanding and mastery of significant texts. An accurate grasp of the causality between events can also provide powerful help for the upper-level application tasks such as chapter comprehension (Ho et al, 2022;Varghese & Punithavalli, 2022), event prediction (Ma et al, 2022;Tomašev et al, 2021;Zhao, 2021), question answering (Das et al, 2022;Jin et al, 2022), and so on. As a more finegrained task than causality extraction, causality detection focuses on the study of sentence-level texts.…”