Ensemble Deep Learning on Time-Series Representation of Tweets for Rumor Detection in Social Media

Chandra Mouli Madhav Kotteti,Xishuang Dong,Lijun Qian

doi:10.3390/app10217541

Chandra Mouli Madhav Kotteti, Xishuang Dong + Show 1 more

Open Access

https://doi.org/10.3390/app10217541

Copy DOI

Abstract

Social media is a popular platform for information sharing. Any piece of information can be spread rapidly across the globe at lightning speed. The biggest challenge for social media platforms like Twitter is how to trust news shared on them when there is no systematic news verification process, which is the case for traditional media. Detecting false information, for example, detection of rumors is a non-trivial task, given the fast-paced social media environment. In this work, we proposed an ensemble model, which performs majority-voting scheme on a collection of predictions of neural networks using time-series vector representation of Twitter data for fast detection of rumors. Experimental results show that proposed neural network models outperformed classical machine learning models in terms of micro F1 score. When compared to our previous works the improvements are 12.5% and 7.9%, respectively.

Highlights

Over the past few decades social media has emerged out as the primary means for news creation as well as for news consumption
We explore the temporal features of Twitter data for timely detection of rumors in social media
We used F1-score, which is the weighted average of Precision and Recall scores as the ensemble model’s evaluation metric

Summary

Introduction

Over the past few decades social media has emerged out as the primary means for news creation as well as for news consumption. The biggest challenge for news spreading on social media is how to verify whether that news is correct or not. Even though social media outperforms traditional media in many aspects, the key difference between them is that the news is verified for its truthfulness before it gets proliferated in traditional media, while it is not the case for social media. Any piece of information can be spread on social media regardless of its truthfulness. Information shared on social media propagates rapidly and increases the difficulty in verifying its credibility in near real time. A rumor is considered as a direct subclass to Pheme, which has four sub-classes. They are speculation, controversy, misinformation, and disinformation

Methods

Results

Conclusion