Skip to content

Latest commit

 

History

History
47 lines (35 loc) · 1.71 KB

README.md

File metadata and controls

47 lines (35 loc) · 1.71 KB
Raw

online-news-popularity-prediction

This is a project for Machine Learning college course, which was implemented by a two person team consisting of @JakubDralus and @Veczar.
It aims to predict the popularity of an article using random forest classifier.

Note: we have taken various appraches to this problem as seen in model.ipynb file. We later discovered that random forest classifier to be the right model to use.

Technologies

  • Python 3.11.5
  • Conda 23.10.00
  • pandas
  • numpy
  • sklearn
  • matplotlib
  • seaborn

Dataset

The dataset has 59 features and a target which is the number of shares in social networks (popularity).
This is how it looks afret scailing:

image

Results

We managed to get an accuracy of 0.67 using random forest classifier. In dataset description it is a recommended best model.

Accuracy: 0.67
              precision    recall  f1-score   support

           0       0.66      0.60      0.63      5591
           1       0.67      0.73      0.70      6303

    accuracy                           0.67     11894
   macro avg       0.67      0.67      0.67     11894
weighted avg       0.67      0.67      0.67     11894

The Dataset Citation

https://archive.ics.uci.edu/dataset/332/online+news+popularity

K. Fernandes, P. Vinagre and P. Cortez. A Proactive Intelligent Decision Support System for Predicting the Popularity of Online News. Proceedings of the 17th EPIA 2015 - Portuguese Conference on Artificial Intelligence, September, Coimbra, Portugal.