在社交媒體的快速發展下，伴隨著假新聞氾濫問題日益嚴重，進而加劇了社會的極化 現象。在這樣的情況下使得檢測假新聞的議題愈發重要。然而假新聞的迅速傳播和演 變，以及缺乏全面的數據，帶給假新聞檢測很大的難關。以前使用不同方法檢測假新 聞的嘗試，在準確率方面遇到了限制。為了應對這些挑戰，我們提出了一種集成文本 和社交網絡信息的多模式假新聞檢測 (MFND) 方法。 MFND 結合了新聞的語義表示、 社交媒體上的傳播模式和超圖技術來豐富數據集。 在兩個真實世界數據集上的實驗結 果表明，MFND 優於最先進的模型，實現了更高的準確性。;With the rapid development of social media, the proliferation of fake news has become a pressing issue, further exacerbating societal polarization. Detecting fake news has become increasingly important in such circumstances. However, the rapid spread and evolution of fake news, along with the lack of comprehensive data, pose significant challenges to fake news detection. Previous attempts to detect fake news using different methods encountered limitations in terms of accuracy. To address these challenges, we propose a Multimodal Fake News Detection (MFND) approach that integrates text and social network information. MFND combines semantic representation of news, propagation patterns on social media, and hypergraph techniques to enrich the dataset. Experimental results on two real-world datasets demonstrate that MFND outperforms state-of-the-art models, achieving the highest accuracy on fake news detection.
