Personalized high quality news recommendations using word embeddings and text classification models

EasyChair Preprint 1254

Abstract

Reading news articles is one of the most important activities online and many apps have appeared the last few years for this purpose. In this paper, we present the architecture of a news recommendation system that provides personalized results to the users. We introduce a method to model the users’ interests over time using word embeddings and a framework to filter and score high quality news stories using text classification models and agglomerative news clustering.

Keyphrases: News Recommendation, Recommender Systems, Recurrent Neural Networks, text classification, word embedding

@booklet{EasyChair:1254,
  author    = {Christos Samarinas and Stefanos Zafeiriou},
  title     = {Personalized high quality news recommendations using word embeddings and text classification models},
  howpublished = {EasyChair Preprint 1254},
  year      = {EasyChair, 2019}}