6th International Workshop on News Recommendation and Analytics (INRA 2018) Özlem Özgöbek1 , Benjamin Kille2 , Jon Atle Gulla1 , Martha Larson3 , Andreas Lommatzsch2 1 NTNU, Trondheim, Norway, {ozlem.ozgobek,jon.atle.gulle}@ntnu.no 2 Berlin Intitute of Technology, Berlin, Germany, {firstname.lastname}@dai-labor.de 3 Radboud University, Nijmegen, The Netherlands, m.larson@cs.ru.nl news and disinformation, user experience, and evalua- tion. News recommendation attempts to identify the Abstract most relevant news articles for individual readers. Nu- merous directions promise advancement in the state of Publishers inform citizens about important the art, including context-awareness, representing ar- events. Digitalization overwhelms readers ticles by word embeddings, or gamification to collect with information. Publishers employ rec- more data. Besides, news recommender systems ought ommender systems to tackle challenges re- to address scalability due to issues of big data, diver- lated information access. This workshop series sity, and mobility. News analytics concerns learning invites researchers, practitioners, and other more about news reading behavior. For instance, re- stakeholders to exchange ideas about how searchers investigate social media usage or sentiment. to deal with news. INRA’s sixth edition Fake news and deliberate disinformation have emerged highlights aspects of news recommendation as significant issues. Currently, operators of media- and analytics including innovative algorithms, related platform struggle to determine the trustwor- trust, and misinformation, and privacy. thiness of news. User experience affects the long-term success of publishers’ services. Privacy and data au- thority have become central demands of customers. 1 Scope and Goals Researchers and businesses have yet to establish stan- Media companies turn data into a valuable commod- dardized evaluation protocols. Publicly available data ity. Further, industries, as well as political organiza- sets and platforms seek to support the advancement tions, depend critically on high-quality information in of standardized evaluation, which will in turn advance order to operate effectively. Both groups of stakehold- trustworthy new recommendation technology. ers must deal with the limited amount of time and ef- INRA offers a venue for researchers, practitioners, fort that consumers are willing to devote to processing and interested parties to come together. The primary vast amounts of information. To connect consumers objectives are support of the exchange of ideas and with content in the face of this challenge, automatic promotion of innovation. At the workshop, researchers systems are used to filter information thus facilitating have the opportunity to present findings and proto- information access. In the case of news media, news types. Practitioners can share their everyday prob- recommender systems rely on sophisticated data ana- lems. Media critics can elaborate on societal effects lytics to capture reading interests. Modeling readers’ and issues of news consumption. Additionally, invited changing concerns remains an area of active research. experts present their insights on recent trends. The INRA workshop’s sixth edition is co-located with International Conference on Information and Knowledge Management (CIKM) 2018. At this work- shop, contributors and attendees focus on issues in- 2 Workshop Details cluding news recommendation, news analytics, fake Copyright © CIKM 2018 for the individual papers by the papers' The full-day workshop features seven paper presenta- authors. Copyright © CIKM 2018 for the volume as a collection tions, three keynote speeches, and an interactive ses- by its editors. This volume and its papers are published under sion for demonstrations and discussion. the Creative Commons License Attribution 4.0 International (CC BY 4.0). 2.1 Accepted Papers 3 Workshop Organizers INRA 2018 have received nine submissions. The fol- Özlem Özgöbek (NTNU, Trondheim, Norway) lowing seven papers have been accepted: Özlem works as a postdoctoral researcher at the De- Yao & Hauptmann News Recommendation and Fil- partment of Computer Science at NTNU. Her research ter Bubble. focuses on recommender systems, privacy issues in rec- Chakraborty et al. Predicting Tomorrow’s Headline ommender systems and disinformation detection for using Today’s Twitter Deliberations. online news. She is a co-founder of the Norwegian Big Data Symposium and actively involved in organizing Maddalena, Ceolin & Mizzaro Multidimensional INRA workshop series since 2014. News Quality: A Comparison of Crowd-sourcing and Benjamin Kille (TU Berlin, Germany) Nichesourcing. Benjamin works as a researcher at TU Berlin. He Lim, Jatowt & Yoshikawa Understanding Charac- organized multiple data-driven competitions including teristics of Biased Sentences in News Articles. NewsREEL. His PhD thesis focuses on context-aware Khattar et al. RARE: A Recurrent Attentive Rec- news recommender systems. ommendation Engine for News Aggregators. Jon Atle Gulla (NTNU, Trondheim, Norway) Ollikainen Clustering Enhancement for a Token- Jon Atle is a professor of Information Systems at the Based Recommender. Department of Computer Science in Trondheim. His Kille, Özgöbek & Lommatzsch Defining Contex- research centers on the semantic web, information re- tual Factors for News Consumption. trieval, recommender systems, text analytics, and big data. He is the co-founder of Norwegian Big Data 2.2 Keynote Speakers Symposium and INRA Workshop series. Martha Larson (Radboud University, Nijmegen) The workshop has features three keynote speeches: Martha holds professorships at Radboud University Frank Hopfgartner (Information School of Univer- and TU Delft. Her research focuses on multimedia sity of Sheffield, UK) News Retrieval and Recommen- computing, language technology, information retrieval, dation Initiatives. and recommender systems. Martha has contributed to Anja Benner-Tischler (University of Kassel, Ger- the organization of a variety of workshops and bench- many) EU Data Protection Law and its Influence on marks including MediaEval and NewsREEL. Webtracking–Everything new with the General Data Andreas Lommatzsch (TU Berlin, Germany) Protection Regulation and the ePrivacy-Regulation? Andreas Lommatzsch works as a senior researcher and Leif Ramming (plista GmbH, Germany) Generating director of the Application Center “Data Analytics” at Recommendations at Scale the DAI-Lab at the TU Berlin. He conducts research in the domains of distributed knowledge management and machine learning. He co-organizes the NewsREEL 2.3 Previous Workshops challenge focusing on recommender algorithms for on- INRA’17 held in conjunction with IEEE/WIC/ACM line news portals. International Conference on Web Intelligence (23 Au- gust, Leipzig, Germany)1 4 Program Committee Members INRA’16 held in conjunction with the 24th Confer- Andreas Lommatzsch, Berlin Institute of Technology ence on User Modeling, Adaptation and Personaliza- Asbjørn Følstad, SINTEF, Norway tion (16 July, Halifax, Canada)2 Benjamin Kille, Berlin Institute of Technology INRA’15 held in conjunction with ACM RecSys (20 Cristina Marco, NTNU, Norway September, Vienna, Austria)3 Dietmar Jannach, Unversität Klagenfurt, Austria NRA’14 held in conjunction with the 22nd Confer- Frank Hopfgartner, University Sheffield, UK ence on User Modeling, Adaptation and Personaliza- Humberto Castejon, Telenor Group, Norway tion (11 July, Aalborg, Denmark)4 Jon Atle Gulla, NTNU, Norway NRS’13 held as International News Recommender Lemei Zhang, NTNU, Norway Systems Workshop and Challenge (NRS) in conjunc- Martha Larson, Radboud University Nijmegen tion with ACM RecSys (13 October, Hong-Kong)5 Özlem Özgöbek, NTNU, Norway Peng Liu, NTNU, Norway 1 http://research.idi.ntnu.no/inra/2017/ Toon De Pessimier, imec-UGent-Waves, Ghent Unver- 2 http://research.idi.ntnu.no/inra/2016/ 3 http://research.idi.ntnu.no/inra/2015/ sity, Belgium 4 http://research.idi.ntnu.no/nra2014 Ville Ollikainen, VTT Technical Research Centre, Fin- 5 http://recsys.acm.org/recsys13/nrs land