Skip to content

Creates a training set and uses supervised learning to build a model which predicts whether a webpage is relevant or not relevant for the PD project.

Notifications You must be signed in to change notification settings

PlanetaryDefense/PD-Webpage-Classifier

 
 

Repository files navigation

PD-Webpage-Classifier

Creates a training set and uses supervised learning (text classification) to build a model which predicts whether a webpage is relevant or not relevant based on features extracted from the website URL and title of the page.

Notes on Current Training Set

The current training set includes 1271 total samples, 700 of which are relevant and 571 of which are not relevant. This training set currently yields an F1 score of .84 for the "relevant" class.

About

Creates a training set and uses supervised learning to build a model which predicts whether a webpage is relevant or not relevant for the PD project.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%