News is generating renewed interest in IR prompted by the recent seismic shifts in the global newspaper industry and the changes in audience habits to consume news [1]. While printed news still makes up the better part of the global newspaper revenues, digital news is a fast growing contributor, especially in North America and Europe.
Consumer and reader behaviour is changing at an equally high pace. Since with digital publishing the constraint on publishing deadlines and morning printing became obsolete, digital news is now consumed by more, both in the mornings and evenings [2], while there is also a constant flow of regular users throughout the day. At the same time, the way people are reading news online is also changing. Desktop usage is becoming dwarfed by reading on tablets and mobiles. According to [5], 75% of readers with smartphones and 70% with tablets check the news more than once a day. They read news on the move and in different contexts. A similar statistic by the Media Insight project claims that 69% of millennials get news at least once a day [4].
While a large proportion of this traffic still lands on news portals on the Web, i.e., desktop users and mobile users through mobile browsers, an increasing volume of users consume news via dedicated news apps or through social media.
Social media has affected considerably how people read news, blogs and consume media in general [3]. Social platforms, like Twitter, Facebook, Reddit, Quora and others, allow users to share content and read content shared by their friends. Sharing in this context provides a form of verification of the content, the sharer’s confirmation that is worth reading. This form of social news consumption is especially prevalent in the behaviours of millennials, who tend not to read news in discrete sessions or by going directly to news providers. Instead, news and information are woven into an often continuous experience, which mixes news with social connection, problem solving, social action, and entertainment [4]. Sharing also appears to be draw users into news that they might otherwise have ignored without their peers recommendation and contextualisation. On the other hand, research also reveals that encountering news through social media is not strictly passive or random. People actively navigate and make choices about which sources in their social media feeds they consider to be reliable, while they also actively participate in the dissemination and contextualisation of news, e.g., commenting, liking, favouriting and sharing forward.
A survey by Mobiles Republic, a global news syndication company, indicated that as the number of news outlets grows, so do readers’ appetites for accurate, multi-sourced and fresh news [5], which is resulting in a behaviour called “news snacking”, whereby users check new sources more frequently but spend less time per session. According to the study, the use of branded news applications is declining, while news aggregators are on the rise.
The increasing use of mobiles for rich media consumption and the growing popularity of news is reflected in the growing number of news apps available and their growing user base. For example, Flipboard is a popular news aggregator that provides a personalised news magazine and boosts over 100 million installs. Other popular apps, such as BuzzFeed, Yahoo News, Feedly, StumbleUpon and of course the new Apple News are but the top of the ice berg. The plethora of news portals and news apps is a clear sign of users’ hunger for news consumption but also a sign that the problem of how to serve this content to the users is still very much an open problem.
Regardless of whether on the Web or in an app, news publishers and aggregators have to address and overcome a great number of challenges. These include, the verification or assessment of a source’s reliability, the integration of news with other sources of information, such as social media, real-time processing of both news content and social streams, both in multiple languages, different formats and in very high volumes, deduplication, entity disambiguation, automatic summarisation and news recommendation and personalisation, just to mention a few.
Information Retrieval (IR) applied to news has been a popular research area for decades, but it is well overdue a fresh boost of new research in light of the new challenges and the changes that have taken place both in the types and volume of media content available and the way people consume this content. Not only the algorithms to detect events in news, to find related news or to profile a user’s preferences in the sources and topics of news can still be improved but also there are serious gaps in the state of the art that need to be addressed. For example, little is known about aspects of relevance in news or how diversity should be measured, how sentiment or political opinion should be considered, or how social signals, popularity or social context should be incorporated when ranking or recommending news.
In an attempt to spur renewed research into news IR, a new workshop, the First International Workshop on Recent Trends in News Information Retrieval (NewsIR’16), will take place in conjunction with ECIR 2016 in Padua, Italy on the 20th of March 2016, see http://research.signalmedia.co/newsir16/.
The workshop aims to stimulate discussion around new and powerful uses of IR applied to news sources. Contributions on any of the multiple IR tasks are invited to help solve real user problems in this area.
To accompany the workshop, the organisers released a new data collection suitable for many research projects. The corpus consists of around one million recent news articles from a wide range of sources, see http://research.signalmedia.co/newsir16/signal-dataset.html.
One goal of the workshop is to define shared challenges using this data, such as news recommendation, deduplication, multi-document summarisation, event detection and clustering.
Submissions of applied research project that makes use of this data are encouraged, although this is not required. Potential retrieval tasks that can be studied with this data include:
● detecting and summarising events over time;
● identifying bias in news sources to different topics and/or different entities;
● identifying influencers in media coverage and visualising information flow.
References
[2] http://blog.visiolink.com/are-newspaper-reading-habits-changing
[4] http://www.americanpressinstitute.org/publications/reports/survey-research/millennials-news/