An assistant for Wikipedia administrators that predicts the protection level and protection duration as per Wikipedia's page protection policy.

Using state-of-the-art machine learning techniques:

1. Predict protection_level of a given Wikipedia article
2. Predict the protection_duration for the said article

1. Install all dependencies mentioned in the requirements.txt file
2. Run the scripts/get_titles.py file to fetch all Wikipedia titles
3. Run the scripts/get_data.py file to fetch all data for a given title
4. Run: $python .\scripts\get_features_all_pages.py .\dataset\trial_unbalanced\titles_unprotected.csv .\dataset\trial_unbalanced\collected_features.csv 0 to get all features. The second parameter is the input csv file containing
5. Run the scripts/make_predictions.py file to apply machine learning algorithms and get predictions.
6. Demo.py instantiates the Gradio user interface and hosts a local webapp.

1. Abhijeet Lokhande

1. DePP 2016
2. Detecting pages to protect in Wikipedia across multiple languages 2019