/News-Recommendation-System

News-Recommendation-System-using-SVM

Primary LanguageJupyter Notebook

News-Recommendation-System-using-Content-based-Filtering

Introduction

Finding relevant and interesting news from a large corpus of news can be a difficult, time-consuming and irritating process.

Given a user with his/her previously read articles, we would like to recommend new articles from the dataset which will be relevant to the user.

Machine Learning Algorithms :

  • SVM
  • Cosine Similarity

Kernels used in SVM :

  • Linear
  • Polynomial
  • Radial

Results

Feature Extraction

alt text

Results for different User

alt text

alt text

alt text

alt text

alt text

alt text

alt text

alt text