natural-language-processing-8

Rumour Detection

3260 papers • 126 benchmarks • 313 datasets

Rumor detection is the task of identifying rumors, i.e. statements whose veracity is not quickly or ever confirmed, in utterances on social media platforms.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in rumour-detection-8

Trend

Dataset

Best Model

Actions

Sepehr_RumTel01

Libraries

i

Use these libraries to find rumour-detection-8 models and implementations

Datasets

Sepehr_RumTel01

FTR-18

CIDII Dataset

Subtasks

No subtasks available.

Most implemented papers

Learning Reporting Dynamics during Breaking News for Rumour Detection in Social Media

A. Zubiaga, Maria Liakata, R. Procter•Sun Oct 23 2016

A novel approach to rumour detection that learns from the sequential dynamics of reporting during breaking news in social media to detect rumours in new stories and achieves competitive performance, beating the state-of-the-art classifier that relies on querying tweets with improved precision and recall, as well as outperforming the best baseline.

194

Content

0

Paper Graph

Turing at SemEval-2017 Task 8: Sequential Approach to Rumour Stance Classification with Branch-LSTM

Isabelle Augenstein, Maria Liakata, E. Kochkina•Sun Apr 02 2017

A LSTM-based sequential model is proposed that, through modelling the conversational structure of tweets, achieves an accuracy of 0.784 on the RumourEval test set outperforming all other systems in Subtask A.

143 0

Paper Graph

The Surprising Performance of Simple Baselines for Misinformation Detection

Kellin Pelrine, Jacob Danovitch, Reihaneh Rabbany•Tue Apr 13 2021

This paper examines the performance of a broad set of modern transformer-based language models and shows that with basic fine-tuning, these models are competitive with and can even significantly outperform recently proposed state-of-the-art methods.

73 0

Paper Graph

Rumor Detection on Twitter with Tree-structured Recursive Neural Networks

Jing Ma, Wei Gao, Kam-Fai Wong•Sat Jun 30 2018

This work proposes two recursive neural models based on a bottom-up and a top-down tree-structured neural networks for rumor representation learning and classification, which naturally conform to the propagation layout of tweets.

550 0

Paper Graph

Stance Classification for Rumour Analysis in Twitter: Exploiting Affective Information and Conversation Structure

Valerio Basile, V. Patti, Endang Wahyu Pamungkas•Sun Jan 06 2019

A new approach to this task, where the use of conversation-based and affective-based features, covering different facets of affect, has been explored, showing the effectiveness of the feature set proposed.

38 0

Paper Graph

BUT-FIT at SemEval-2019 Task 7: Determining the Rumour Stance with Pre-Trained Deep Bidirectional Transformers

Martin Fajcik, P. Smrz, Luk'avs Burget•Sun Feb 24 2019

This paper describes the system submitted to SemEval 2019, classifying whether posts from Twitter and Reddit support, deny, query, or comment a hidden rumour, truthfulness of which is the topic of an underlying discussion thread.

45 0

Paper Graph

CLEARumor at SemEval-2019 Task 7: ConvoLving ELMo Against Rumors

Steffen Staab, Ipek Baris Schlicht, Lukas Schmelzeisen•Thu Apr 04 2019

This paper describes the submission to SemEval-2019 Task 7: RumourEval: Determining Rumor Veracity and Support for Rumors, and provides results and analysis of the system performance and present ablation experiments.

16 0

Paper Graph

Danish Stance Classification and Rumour Resolution

Anders Edelbo Lillie, Emil Refsgaard Middelboe•Mon Jul 01 2019

This thesis generates a stance-annotated Reddit dataset for the Danish language, and implements various models for stance classification, which shows that stance labels can be used across languages and platforms with a HMM to predict the veracity of rumours.

2 0

Paper Graph

Claim Check-Worthiness Detection as Positive Unlabelled Learning

Isabelle Augenstein, Dustin Wright•Wed Mar 04 2020

The best performing method is a unified approach which automatically corrects for this using a variant of positive unlabelled learning that finds instances which were incorrectly labelled as not check-worthy.

47 0

Paper Graph

Adding a benchmark result helps the community track progress.