2nd Norwegian Big Data Symposium will be held in Trondheim in November 2016. For more details: NOBIDS
Our paper on usability factors of news recommender systems is accepted to SMAP 2016 : “Interactive Mobile News Recommender System: A Preliminary Study of Usability Factors ” by Xiaomeng Su , Özlem Özgöbek, Jon Atle Gulla, Jon Espen Ingvaldsen, Arne Dag Fidjestøl.
Paper in CICLing 2016
Our paper "Hybrid Entity Driven News Detection on Twitter" by Linn Vikre, Henning Wold, Özlem Özgöbek and Jon Atle Gulla is accepted for CICLing 2016 conference.
INRA 2016 in UMAP 2016
4th International Workshop on News Recommendation and Analytics (INRA 2016) will be held in conjunction with UMAP 2016. Details are coming soon!
Mobile news recommender systems help users retrieve news that is relevant in their particular context and can be presented in ways that require minimal user interaction. In spite of the availability of contextual information about mobile users, though, current mobile news applications employ rather simple strategies for news recommendation. Our multi-perspective approach unifies temporal, locational, and preferential information to provide a more fine grained recommendation strategy.
The NTNU Smartmedia program at the Department of Computer and Information Science, Norwegian University of Science and Technology, was established in 2012 in close collaboration with the Scandinavian media industry. As the industry is addressing the abundance of news and information in general from news agencies and social sites, as well as open data from public and private institutions, it has become paramount to develop architectures and technologies for large-scale realtime data processing. The intention of Smartmedia is to look into new technologies that may help the companies and their journalists deal with the explosion of online information and present news more efficiently and attractively to readers. Central to this program are technologies like:
Big Data architectures
Information retrieval and recommendation
Text analytics and sentiment analysis
Currently, we run the project REX - Recommendation Technologies for Online News Streams. REX receives the financial support from Research Council of Norway, Adresseavisen ASA and Cxense with a total budget around 4.7 million Euro.
The main objective in the REX project is to develop next generation's news recommendation for online news streams. The approach in REX combines news articles' underlying semantics with large-scale analysis of heterogeneous user data. Linked open data is used to disambiguate and enrich entities recognized in news articles and relate them to extensive user profiles that are continuously updated on the basis of users' interaction with the system. One of the largest media houses in Norway, Polaris Media, is the project owner, while most of the work on the recommender system itself is done by Cxense, a software company developing and selling user profiling and news recommendation services to the media industry all over the world.