My project : "FAKE NEWS DETECTION" USING BERT with CNN.
In this project, we propose a BERT-based (Bidirectional Encoder Representations from Transformers) deep learning approach by combining different parallel blocks of the CNN layer and deep Convolutional Neural Network (CNN) having different kernel sizes and filters with the BERT. This work proposes a hybrid deep learning model that combines convolutional Neural Network and BERT for fake news classification. The model was successfully validated on LIARPLUS.
LIAR-PLUS dataset having news data of six categories. In this model, we used Convolutional neural network with two dimensional layers and BERT encoder for model training and also obtained most promising results.
LIAR-PLUS dataset having news data of six categories:
Here the dataset consists of six labels (Pants-fire, Barely-True, False, Half-True, Barely-True, True) where each label is considered as a separated class.