Fake_News_Detection
All data we use could be accessed at https://www.kaggle.com/c/fake-news/data.
Before running the code, make sure python3 is used. The required libraries are numpy, panda matplotlib, gensim sklearn and nltk. Run pip install to install the missing packages for some of the deep learning models.
To start, download the data and make a new directory named datasets or customize the path in driver.py.
To preprocess the data, run driver to generate xte.npy,yte.npy,xtr.npy,ytr.npy and get the results for naive bayes and SVM for our baseline.
To obtain the data analytics insight on our data, run jupyter notebook to execute data_analysis.ipynb.
Use jupyter notebook to execute DenseNet.ipynb to view results for Neural network, LSTM.ipynb to view results for LSTM and BiLSTM.ipynb to view results for BiLSTM.