FakeNewsDetection

Fake news, or the online sharing of factually incorrect articles, has emerged as a consequence of the widespread adoption of social media as a way to pass news to others. The main task of the project is to solve the problem of fake news with the help of technologies such as deep learning and NLP. The task of fake news detection is difficult as compared to the task of image recognition and text translation and classification as it needs to know the right context of the text and chances of mistakes are really high. Normal Deep learning architectures do not work so well in the task the main goal of the project is to make a custom architecture to facilitate the task of fake news detection.

The project will be done in 5 steps-

1.Finding the dataset

2.Understanding the dataset and analyzing and cleaning it

3.Applying current available architectures

4.Finding new architures for the task

5.Presenting the results

Libraries Used

I will make use of python as a primary language for the task and for deep learning i will be using frameworks such as PyTorch and Keras(backend TensorFlow) for the construction of model and cloud technologies like AWS or Azure can be used for training the model. I can also use Flask to make api if i get enough accuracy and time but it is not the final goal of the project.

Existing architectures or models , I will try for the problem are-

1.RNN’s

2.CNN’s

3.GRU’s

4.LSTM

5.AutoEncoders

6.Ensemble Models etc.

And finally different ways to represent words in NLP like Word2Vec,GlobalVec etc.