Learn over Past, Evolve for Future: Forecasting Temporal Trends for Fake News Detection
This work addresses the temporal shift issue in fake news detection, which is critical for maintaining online news ecosystem health, but it is incremental as it builds on existing detection methods by incorporating temporal forecasting.
The paper tackles the problem of fake news detection by addressing temporal shifts in news data, proposing a framework that forecasts temporal distribution patterns to adapt models to future data, achieving superior performance on a real-world dataset.
Fake news detection has been a critical task for maintaining the health of the online news ecosystem. However, very few existing works consider the temporal shift issue caused by the rapidly-evolving nature of news data in practice, resulting in significant performance degradation when training on past data and testing on future data. In this paper, we observe that the appearances of news events on the same topic may display discernible patterns over time, and posit that such patterns can assist in selecting training instances that could make the model adapt better to future data. Specifically, we design an effective framework FTT (Forecasting Temporal Trends), which could forecast the temporal distribution patterns of news data and then guide the detector to fast adapt to future distribution. Experiments on the real-world temporally split dataset demonstrate the superiority of our proposed framework. The code is available at https://github.com/ICTMCG/FTT-ACL23.