Attention-Based LSTM Network for Rumor Veracity Estimation of Tweets

Abstract

Twitter has become a fertile place for rumors, as information can spread to a large number of people immediately. Rumors can mislead public opinion, weaken social order, decrease the legitimacy of government, and lead to a significant threat to social stability. Therefore, timely detection and debunking rumor are urgently needed. In this work, we proposed an Attention-based Long-Short Term Memory (LSTM) network that uses tweet text with thirteen different linguistic and user features to distinguish rumor and non-rumor tweets. The performance of the proposed Attention-based LSTM model is compared with several conventional machine and deep learning models. The proposed Attention-based LSTM model achieved an F 1-score of 0.88 in classifying rumor and non-rumor tweets, which is better than the state-of-the-art results. The proposed system can reduce the impact of rumors on society and weaken the loss of life, money, and build the firm trust of users with social media platforms.

A Confirmation Bias View on Social Media Induced Polarisation During Covid-19

“…The proliferation of social media has been a catalyst to inducing a dangerous, socio-cultural polarisation of society (Montalvo & Reynal-Querol, 2005 ; Spohr, 2017 ; Vishwanath, 2015 ). This polarisation phenomenon has led researchers to explore SMIP from different theoretical lenses to study specific social media platforms and particular attention given to fake news (Langley et al, 2021 ; Brummette et al, 2018 ; Lee et al, 2015a ; Shearer & Grieco, 2019 ; Singh et al, 2020 ). Manifestations of SMIP include misinformation about presidential election campaigns (Guess et al, 2018 ; Linvill & Warren, 2020 ; Schäfer et al, 2017 ), race (Jamieson, 2020 ), immigration (Jaramillo-Dent & Pérez-Rodríguez, 2021 ; Newman et al, 2017 ), religion (Said, 2008 ), and pandemics, specifically Covid-19 (Laato et al, 2020 ).…”

Fake news detection using naïve Bayes and long short term memory algorithms

“…In machine learning, a massive number of studies have proposed different models of machine learning algorithms like naïve Bayes (NB), support vector machine (SVM), logistic regression (LR), and many other models [2], [8]- [11], Our focus is on naïve Bayes model, many fake news detections have been implemented using NB as in the second model of the work done by [12], NB achieved 56% accuracy among 7 other classifiers using linguistic and user features. The work [13] shows the model NB classified the second in accuracy after SVM with 55.85%, authors in [14] achieved the highest accuracy of 80% using bayesian classifier.…”

“…Asghar et al [20] suggested a rumor classification model proposed by merging bi-long-short term memory (BiLSTM) with CNN. Moreover, Singh et al [12] proposed an attention-based LSTM network that distinguishes rumor and non-rumor tweets using tweet text and 30 different linguistic and user features. The performance of the model reached 88% against the other conventional machine and deep learning models.…”

MCred: multi-modal message credibility for fake news detection using BERT and CNN

“…Singh et al. ( 2020 ) used attention-based LSTM to classify rumor and non-rumor tweets with thirteen linguistic and user profile features and achieved an F1-score of 88%. Horne and Adali ( 2017 ) developed support vector machine (SVM) model for fake news detection using three linguistics features categories: writing pattern, text complexity, and psychological.…”