Al Jazeera In 1000 Academic Studies

138. Name: Youssef Meguebli Title: Leveraging User-Generated Content for Enhancing and personalizing News Recommendation (CNN, Al Jazeera, The

Independent, and The Telegraph) Institution: Université Paris-Saclay Country: France Date: 2015 Language: English Abstract:

Online news websites are becoming very popular and influential social media platforms allowing people to easily access information and give feedback. This study investigates how to exploit user-generated content for personalized recommendations. Opinions provided by users represent an important indicator about their profiles. By mining such content, we can extract valuable information about their interests, their inclination towards particular articles, their political orientation, their favorite sport teams and many other interesting information. In the first part, the researcher developed a fine-grained model that captures both users and article profiles. The profile of each user is extracted from all the opinions and reactions that are provided on the news websites, while the profile of an article is extracted from its content. The accuracy of the extracted profiles is assessed on two datasets crawled from CNN and Al Jazeera and the results show high quality results. It was also assessed on a large test collection based on real users’ activities in four news websites: The Independent, The Telegraph, CNN and Al Jazeera and the results show that this model outperforms baseline models, achieving high accuracy.


Made with FlippingBook Online newsletter