3rd International Workshop on News Recommendation and Analytics (INRA 2015) Jon Atle Gulla Bei Yu Özlem Özgöbek Department of Computer and School of Information Studies Computer Engineering Information Science Syracuse University, USA Department NTNU byu@syr.edu Balikesir University Trondheim, Norway Balikesir, Turkey jag@idi.ntnu.no ozlemozgobek@gmail.com Nafiseh Shabib Department of Computer and Information Science NTNU Trondheim, Norway shabib@idi.ntnu.no ABSTRACT mender systems have particular challenges which requires The 3rd International Workshop on News Recommendation a deeper analysis of both the user, content and their rela- and Analytics (INRA 2015) is held in conjunction with Rec- tionships. The news domain is characterized by a constant Sys 2015 Conference in Vienna, Austria. This paper presents flow of unstructured, fragmentary, and unreliable news sto- a brief summary of the INRA 2015. This workshop aims to ries from numerous sources and different perspectives. Some create an interdisciplinary community that addresses design important challenges of news domain are: issues in news recommender systems and news analytics, • Dynamic environment: Every hour hundreds of new and promote fruitful collaboration opportunities between re- articles is published by different sources, searchers, media companies and practitioners. We have a keynote speaker and an invited demo presentation in addi- • Faster changing user interests compared to other do- tion to 4 papers accepted in this workshop. mains. User interests in movies, music or books change much slower than news, Categories and Subject Descriptors • Willingness to read news articles that are independent H.0 [Information Systems]: General from user interests like breaking news, • Recency issues of news articles (people tend to read Keywords recent news, not the old ones), Recommender systems; news recommendation; analytics • Unstructured subjective content that create content analysis problems and may turn recommendations un- 1. INTRODUCTION reliable. The motivation for news recommender systems is the tremen- This workshop addresses primarily news recommender sys- dous amount of news articles available online and the dy- tems and news analytics, with a particular focus on user pro- namic nature of news domain. For a user it is getting harder filing and techniques for dealing with and extracting knowl- to reach the relevant news items according to her personal edge from large-scale news streams. The news streams may interests and preferences. News recommender systems aim originate in large media companies, but may also come from to bring the most relevant news items to the users. social sites, where user models are needed to decide how Each domain in recommender systems has different char- user-generated content is to be taken into account. As part acteristics and requires different approaches to make suc- of news recommendation and analytics, Big Data architec- cessful recommendations. Compared to other recommender tures and large-scale statistical and linguistic techniques are system domains like books, music and movies, news recom- used to extract aggregated knowledge from large news streams and prepare for personalized access to news. 2. TOPICS OF INTEREST Topics of interests for this workshop include but are not limited to: Copyright is held by the owner/author(s). • News semantics and ontologies, INRA 2015 in conjunction with RecSys ’15, September 16-20, 2015, Vienna, Austria • News summarization, classification and sentiment anal- . ysis, • Recommender systems and news personalization, • International News Recommender Systems Workshop and Challenge (NRS)3 held in conjunction with the 7th • Group recommendation for news, ACM Recommender Systems Conference in 2013. This workshop had a very limited scope, which restricted • User profiling and news context modeling, the number of submissions and led to an acceptance • News evolution and trends, rate of 75%. • Large-scale news mining and analytics, • International Workshop on News Recommendation and Analytics (NRA) 2014 4 held in conjunction with 22nd • Evaluation methods, Conference on User Modelling, Adaptation and Per- sonalization (UMAP) in 2014. The workshop scope • News from social media, was extended with news analytics, which is closely lined with the field of news recommendation. The ac- • Big Data technologies for news streams, ceptance rate was 50%. • News recommendation and analytics on mobile plat- forms. 4. ORGANIZERS 4.1 Workshop Chairs 3. WORKSHOP DETAILS Jon Atle Gulla, Professor at Department of Computer In INRA 2015 we have received 6 submissions, of which and Information Science, Norwegian University of Science 4 were accepted for presentation. The submissions to our and Technology (NTNU), Norway workshop includes good quality of works for user profiling, social network analysis, hybrid methods and deep learning Bei Yu, Assistant Professor at School of Information for news recommendation. This year we have the acceptance Studies, Syracuse University, USA rate of 66%. In INRA 2015 we have a keynote speaker who has a very good background of industrial and academic work 4.2 Organizing Committee Co-Chairs on data analytics and news research. INRA also includes a Özlem Özgöbek, Computer Engineering Department, demo session with a full-fledged news recommender system Balikesir University, Turkey that has been developed as part of the SmartMedia program at NTNU in Trondheim. This system integrates numerous Xiaomeng Su, Department of Informatics and eLearn- Norwegian news sources and demonstrates how the use of ing, Sør-Trøndelag University College, Norway context can improve the quality of personalized mobile news aggregators. 4.3 Program Committee Co-Chairs 3.1 Keynote Speaker Nafiseh Shabib, Norwegian University of Science and Dr. Bahareh Heravi is the keynote speaker of INRA 2015. Technology (NTNU), Department of Computer and Infor- She is a Research Fellow at Insight Centre for Data Analytics mation Science, Norway @ NUI Galway and the founder and Leader of the Insight News Lab 1 (Digital Humanities and Journalism). She is Jon Espen Ingvaldsen, Norwegian University of Science an Adjunct Lecturer and a programme board member of and Technology (NTNU), Department of Computer and In- Journalism Studies at the National University of Ireland, formation Science, Norway Galway. Dr Heravi’s research and teaching are at the nexus of data, technology and journalism. She has worked with and within various news organizations. Her main research interests are in the areas of Data and Computational Jour- nalism, Information Science, Social Media & Citizen Sens- ing, Semantic Web and Linked Data. She leads a number of projects at the Insight News Lab, including RTE News360, Newswire 2.0, VeritaWire, Social Semantic Journalism and the Social Repository of Ireland. Dr Heravi is the founder of Irish Times Data 2 and also the founder and organizer of Hacks/Hackers Dublin. She is the former Lead Data Scien- tist at The Irish Times, where she lead the Irish Times’R&D programme and pioneered data innovation in the newsroom. Dr. Heravi’s talk is about data journalism as a practice, the new directions and challenges in news organizations. 3.2 Previous Workshops 3rd International Workshop on News Recommendation and Analytics (INRA 2015) is based on the following previ- ous workshops: 1 3 http://newslab.insight-centre.org/ http://recsys.acm.org/recsys13/nrs 2 4 http://irishtimes.com/data http://research.idi.ntnu.no/nra2014