A vertical news recommendation system: CCNS—An example from Chinese campus news reading system

Abstract

News recommendation systems are widely used to address the information overloading problem. Many Web-based news reading services, like Google News and Yahoo! News, have become increasingly prevalent as they help users find interesting articles from news providers that match the users' preference. However, few research efforts have been reported on campus news recommendation. Different from news articles, news from vertical systems is often short with limited topic scope, targeting at specific audience. To address the aforementioned characteristics, in the paper, we develop a hybrid recommendation system for campus news by integrating different recommendation algorithms using linear combination. Offline and online experiments are conducted to evaluate the system effectiveness.