关键词: 虚假新闻检测, 领域转移, 软共享内存网络, Transformer, APK-CNN

Abstract: In order to solve the problems of domain shifting and incomplete domain labeling in social media news, as well as to explore more efficient multi-domain news feature extraction and fusion networks, a multi-domain fake news detection model based on APK-CNN (Adaptive Pooling Kernel Convolutional Neural Network) and Transformer enhancement was proposed, namely Transm3. Firstly, a three-channel network was designed for feature extraction and representation of semantic, emotional, and stylistic information of the text and view combination of these features using a multi-granularity cross-domain interactor. Secondly, the news domain labels were refined by optimized Soft-shared Memory Networking and domain adapters, Then, Transformer was combined with a multi-granularity cross-domain interactor to dynamically weight the aggregation of different domain interaction features. Finally, the fused features were fed into the classifier for true/false news discrimination. Experimental results show that compared with M3FEND and EANN, Transm3 improves the macro F1 values by 3.68% and 6.46% on Chinese dataset, and 6.75% and 11.93% on English dataset. And the F1 values on the nine sub-domains are also significantly improved. The effectiveness of Transm3 model for multi-domain fake news detection work is fully validated, providing strong support for fake news detection in social media.

Key words: fake news detection, domain shift, soft-shared memory networking, Transformer, APK-CNN（Adaptive Pooling Kernel Convolutional Neural Network）