Fake News Identification and Stance Detection using CNN-LSTM

This research paper investigates the use of a combination of CNN and LSTM models for identifying fake news and stance detection. Fake news involves analyzing a piece of news article and checking whether the news contains fake or misleading information. Stance detection refers to a writers’ emotion whether he is for or against a particular topic. Stance detection is a complex task which involves NLP and Machine Learning techniques. In the CNN-LSTM model, Firstly- Input layer receives the preprocessed text data which is then passed to the next layer. Secondly, CNN layer, which is used to extract important features from the text. Lastly, the LSTM layer which captures sequential structure of the text. The output of this layer is finally fed to the output layer for binary classification. The performance of the model is evaluated using several metrics like accuracy, precision, recall, and F1 score. Our experimental result demonstrates that the model is highly effective in detecting fake news using both CNN and LSTM. Overall, our research provides important insights into the problem and how the model has the potential to make a significant impact on major issues like “Fake News”. It also has important implications for addressing the challenges of fake news.