Attention-Based Ensemble Deep Learning Model for Arabic and English Fake News Classification
TL;DR
An Ensemble Deep Learning Model used for Arabic and English fake news classification that consists of CNN, Bi-LSTM with attention, and Bi-GRU with attention networks and it can be seen that the combination of all three networks provides additional information that makes the final decision better.
Abstract
It is difficult to classify articles as fake news since one article may consist of true facts with only some statements being fake. Moreover, classification becomes complicated for the Arabic language owing to its morphology and several ways of spelling, as well as the lack of well-classified and marked data sets. This paper presents an Ensemble Deep Learning Model (EDLM) used for Arabic and English fake news classification. The EDLM consists of CNN, Bi-LSTM with attention, and Bi-GRU with attention networks. Each of them produces one probability of the article, which is then summed up to a final probability via a sigmoid classification layer. The presented model was tested using AraNews, the Arabic Fake News Dataset (AFND), and the English Fake-or-Real dataset. Accuracy, precision, recall, and F1-score were used for evaluation. The highest accuracies obtained were 0.9146, 0.8363, and 0.9881, while the F1-scores reached 0.9032, 0.8358, and 0.9881. Compared to the best-performing network under the same conditions, the EDLM improved accuracy by 0.48, 1.80, and 0.74 percentage points correspondingly. It can be seen that the combination of all three networks provides additional information that makes the final decision better. There are some constraints concerning static embedding, only text input, source-level labels in the AFND, and missing details about the original software environment.