Teammate: Valerios Stais
The goal of this project was to create models able to correctly classify articles belonging in one of the following categories, entertainment, business, health, technology. We focused on extracting as much information as possible with a refined preprocessing of the text. Then we aimed for the best accuracy finding the most appropriate combination of model and hyper-parameters using Bayesian optimization.
For a detailed explanation of the steps we followed to achieve an as good as possible classifier we suggest reading report.pdf.