Predict news article popularity based on text content of the headline and the content.
Following pipeline was designed to combine information from both the headline and the content.
Notebooks/headline.ipynb develops the headline model and adds the output as a feature to the dataset.
Notebooks/demo.ipynb combines the headline model output and constructs a new popularity classifier.