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

2017/04/24 by Elena Kochkina, Maria Liakata, Kochkina, Elena +3
Computer Science · Social Sciences · #Advanced Text Analysis Techniques #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Misinformation and Its Impacts #Topic Modeling

paper · pdf · doi:10.48550/arxiv.1704.07221

openalex publication_date 2017/04/24 · openalex created_date 2022/10/05 · openalex updated_date 2026/07/28

Abstract

This paper describes team Turing's submission to SemEval 2017 RumourEval:\nDetermining rumour veracity and support for rumours (SemEval 2017 Task 8,\nSubtask A). Subtask A addresses the challenge of rumour stance classification,\nwhich involves identifying the attitude of Twitter users towards the\ntruthfulness of the rumour they are discussing. Stance classification is\nconsidered to be an important step towards rumour verification, therefore\nperforming well in this task is expected to be useful in debunking false\nrumours. In this work we classify a set of Twitter posts discussing rumours\ninto either supporting, denying, questioning or commenting on the underlying\nrumours. We propose a LSTM-based sequential model that, through modelling the\nconversational structure of tweets, which achieves an accuracy of 0.784 on the\nRumourEval test set outperforming all other systems in Subtask A.\n

Related

• Stance Classification for Rumour Analysis in Twitter: Exploiting Affective Information and Conversation Structure
• Rumour Evaluation with Very Large Language Models
• Stance Classification in Rumours as a Sequential Task Exploiting the Tree Structure of Social Media Conversations
• Using Gaussian Processes for Rumour Stance Classification in Social Media
• Danish Stance Classification and Rumour Resolution