This repository pertains to the paper "Rumour As an Anomaly Rumour Detection with One-Class Classification", presented at the 2019 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC).
This study addresses the problem of rumour scarcity versus non-rumour abundance in automatic rumour detection. To tackle this issue, we portray rumour as an anomaly by showing how disproportionate is the number of rumours versus non-rumours. This imbalance is scrutinized by comparing the rate of news production versus rate of fact-check production. Then, we exploit one-class classification approach to distinguish rumour from non-rumour. One-class classification separates rumour from non-rumour via training the classifier with only non-rumour. To train the one-class classifier, we extract 33 short-term features, regarding the purpose of this research in early detection of rumours. We evaluate the performance of our model by accuracy and F-score. In terms of F-score, our model outperforms the state-of-the-art and reaches to very close proximity of highest accuracy on the same dataset.
The following figure displays the accuracy and F-score of the proposed model across different feature categories.
To get access to the full paper, please visit paper webpage, or contact the main author.
A. E. Fard, M. Mohammadi, S. Cunningham and B. V. de Walle, "Rumour As an Anomaly: Rumour Detection with One-Class Classification," 2019 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC), 2019, pp. 1-9, doi: 10.1109/ICE.2019.8792644.