{"id":4236,"date":"2016-04-26T11:46:40","date_gmt":"2016-04-26T10:46:40","guid":{"rendered":"https:\/\/irsg.bcs.org\/informer\/?p=4236"},"modified":"2016-04-26T11:46:40","modified_gmt":"2016-04-26T10:46:40","slug":"first-international-workshop-on-recent-trends-in-news-information-retrieval-newsir16","status":"publish","type":"post","link":"https:\/\/archive-irsg.bcs.org\/informer\/?p=4236","title":{"rendered":"First International Workshop on Recent Trends in News Information Retrieval (NewsIR\u201916)"},"content":{"rendered":"
\"\"
News + IR = NewsIR'16 (demonstrated by our distinguished keynote speaker Jochen Leidner)<\/figcaption><\/figure>\n

The First International Workshop on Recent Trends in News Information Retrieval<\/em> (NewsIR\u201916<\/a>) was held\u00a0as part of\u00a0the ECIR 2016<\/a> conference in Padua (Italy) on 20 March 2016. The workshop provided an opportunity for a diverse group of stakeholders\u2014researchers, professionals and practitioners\u2014involved in news-related information to come together and discuss the latest and most powerful uses of IR technology applied to news sources. Interest in the subject, combined with appropriate management and good advertisement, resulted in a fairly successful turn-out: more than 40 attendees were present in the room at all times and for all sessions. To take full advantage of the wide range of perspectives brought by the participants, the presentations were brief and allowed time for questions. Additionally, there were plenty of opportunities for networking during the coffee and lunch breaks, poster and break-out sessions, and the welcome reception at the end of the day.<\/p>\n

 <\/p>\n

<\/p>\n

1\u00a0\u00a0 The Origins of NewsIR\u201916<\/h2>\n

The use of news collections to support IR research has been a common practice for several years. Indeed, the Reuters-21578<\/a> collection, which contained Reuters newswire material, was compiled in 1987. Other collections employed for IR research, such as those in some tracks of the Text REtrieval Conference<\/em> (TREC<\/a>) and the CLEF Initiative<\/a><\/em>, also comprise feeds from news broadcasters and other news-related data. However, it would be a mistake to assume that all problems associated with news search have been successfully \u201csolved\u201d. Moreover, mainstream media outlets are often among the most prominent sources of information\u2014ranging from the influence that newspapers have on elections to the damage to a brand\u2019s reputation that a negative article on a popular blog can cause.<\/p>\n

Realising the importance of news datasets for the IR community, a group of participants who attended ECIR 2015<\/a> in Vienna started to talk about the use of news articles as input to various tasks, ranging from news recommendation to temporal summarisation or real-time clustering. It emerged that there was a good number of researchers interested in news-related IR. Yet, it seemed that the available collections for this kind of research were outdated and usually represented biased versions of the news\u2014because they did not contain enough articles, or only a few sources were considered. Additionally, in some cases the data is cleaned and filtered, as opposed to the noisy entries found in the real word, which creates an unrealistic environment for evaluation. Consequently, Miguel Martinez-Alvarez<\/a>\u2014Co-Founder and Head of Research at Signal Media Ltd<\/a>\u2014set up a Google group<\/a> to continue the discussion on news-related IR after the end of ECIR 2015<\/a>.<\/p>\n

Initially, 55 people joined the Google discussion forum<\/a> and talked about subjects such as news bias, summarisation, clustering, topic classification, entity linking, entity recognition and disambiguation, event detection and social media integration. A couple of questions that permeated the discussion from the very beginning were whether news-related information had been relegated to be \u201cless\u201d important than it really is, and whether it would be worth organising a workshop combining news and IR.<\/p>\n

Eventually, a proposal to start up an international workshop was drafted and submitted to ECIR 2016<\/a>, where it was well received. A total of 9 full papers and 3 short papers were selected by the Programme Committee<\/a> from a total of 19 submissions\u2014each submitted paper was reviewed by at least 3 members of an international reviewing group made of 30 members. Apart from the selected papers, two keynote speakers joined the workshop: Jochen L. Leidner<\/a> (Thomson Reuters<\/a>) and Julio Gonzalo<\/a> (National University of Distance Education<\/a>). The NewsIR\u201916 Workshop website is available here<\/a>.<\/p>\n

2 <\/strong> <\/em> <\/strong> <\/em> The Signal Media One-Million News Articles Dataset<\/h2>\n

To accompany the Workshop, and facilitate conducting research on news articles, Signal Media released a dataset<\/a> intended to serve and encourage the research of the news retrieval community. Such a dataset consists of approximately 1 million news articles from a wide range of sources.<\/p>\n

Originally, the articles of the dataset were gathered by Moreover Technologies<\/a> for a period of 1 month\u2014precisely, 1-30 September 2015. Most of the articles are in English, but there are a few non-English and multi-lingual articles. The sources of the articles include companies such as Reuters<\/a>, and also local news sources and blogs. The number of individual unique sources is over 93k. The dataset contains 265,512 blog articles and 734,488 news articles. The average length of an article is 407.75 words.<\/p>\n

3\u00a0\u00a0 Session 1: Media Monitoring<\/h2>\n

The first session was chaired by Gabriella Kazai (Lumi<\/a>, UK) and included the first keynote, by Jochen L. Leidner<\/a> <\/strong> <\/em>, Director of Research at Thomson Reuters<\/a>. Short descriptions of the keynote and papers presented in this session are listed below.<\/p>\n

Keynote: Recent Advances in Information Access at Thomson Reuters R&D \u2013 News and Beyond<\/h5>\n

By Jochen L. Leidner (Corporate Research & Development, Thomson Reuters<\/a>\u2014London, UK)<\/p>\n

Jochen is currently responsible for the Research and Development at the Thomson Reuters site in London, UK, where approximately 40 scientists and developers are engaged on research activities\u2014strategic forward-looking, and applied and contract research (no product development). As pointed out by Jochen, his keynote was largely an \u201cideas\u201d talk, as opposed to a technical talk.<\/p>\n

\"\"
Live performance at NewsIR'16<\/figcaption><\/figure>\n

Jochen reported briefly on some recent developments made by the Corporate R&D Group at Thomson Reuters. For example, a news recommender system called NewsPlus<\/a><\/em>, and a real-time Twitter rumour detection tool for journalists called REUTERS Tracer<\/a><\/em>. From the area of pharma within IP & Science, Jochen talked about SoMeDoSEs<\/a><\/em>, a pharmaco-vigilance system that uses Twitter to mine adverse events associated with medical drugs, which is used for drug repositioning. From the area of law, Jochen discussed the advanced search engine technology that powers the Westlaw<\/a><\/em> search engine. Finally, in the area of Finance & Risk, the Risk Mining<\/a>, a computer-supported risk register extraction application was introduced.<\/p>\n

According to Jochen, the future of news-related information retrieval lies on the ability to transform news into actionable intelligence. This is critical for proactively preventing and reacting to future events.<\/p>\n

Boolean Queries for News Monitoring: Suggesting New Query Terms to Expert Users<\/h5>\n

By Suzan Verberne (Radboud University<\/a>, the Netherlands), Thymen Wabeke (TNO<\/a>, the Netherlands) and Rianne Kaptein (TNO<\/a> <\/strong> <\/em>, the Netherlands).<\/p>\n

The paper evaluates query suggestions for Boolean queries in a news monitoring system. Users of the system receive news articles that match their queries on a daily basis\u2014but the queries need regular updates, as the news changes continuously.<\/p>\n

Suzan Verberne emphasised during the presentation the importance of tasks that are recall oriented\u2014i.e., tasks where missing a single relevant document is not acceptable. One of the traditional ways to address these tasks is by means of long and complex Boolean queries. This research, however, introduces a method to have candidate query terms suggested from retrieved documents.<\/p>\n

After presenting experimental results and qualitative feedback obtained through a questionnaire answered by the participants in the experiments, the authors conclude that the use of relevance ranking, instead of Boolean retrieval, and a post-filtering mechanism for removing non-relevant terms, will give better user satisfaction.<\/p>\n

Detecting Attention Dominating Moments Across Media Types<\/h5>\n

By Igor Brigadir, Derek Greene and P\u00e1draig Cunningham (Insight Centre for Data Analytics<\/a>, University College Dublin<\/a>).<\/p>\n

This paper focuses on identifying attention dominating moments<\/em> in online media\u2014i.e., moments when everyone seems to be talking about the same issue. To explore attention dominating news stories, three different media sources were studied: mainstream news, blogs and tweets. For the first two sources, the Signal Media dataset<\/a> was used. For the final source, the authors collected a Twitter corpus comprising a large set of newsworthy sources curated by journalists, instead of retrieving tweets based on keywords.<\/p>\n

The paper suggests that it might be possible to identify and track major developments with global impact by linking attention dominating moments across multiple sources on different platforms. Social media communities both influence and are influenced by traditional news media\u2014in fact, stories break simultaneously on both Twitter and traditional news publishers.<\/p>\n

Exploiting News to Categorize Tweets: Quantifying the Impact of Different News Collections<\/h5>\n

By Marco Pavan, Stefano Mizzaro, Matteo Bernardon, Ivan Scagnetto (University of Udine<\/a> – Italy).<\/p>\n

The last paper of the first session was also the recipient of the best paper award of the workshop. This paper is a part of a longer term research that aims at understanding the effectiveness of enriching tweets with information derived from the news, instead of the whole Web as a knowledge source.<\/p>\n

Stefano Mizzaro delivered the presentation and explained how to exploit news articles to enhance tweet categorisation using sets of words extracted from news articles with the same temporal context. Three different features of the news were tested as part of this research: volume<\/em>, variety<\/em> and freshness<\/em>. The experiments confirmed the importance of these three features.<\/p>\n

Future work will look at the impact of the number of documents extracted from the news collection to categorise short texts. There are also plans to investigate which kinds of news is important to consider and which ones are marginal.<\/p>\n

4\u00a0\u00a0 Session 2: News Events<\/h2>\n
Semi-Supervised Events Clustering in News Retrieval<\/h5>\n

By Jack G. Conrad (Thomson Reuters Corporate Research & Development<\/a>, Minnesota \u2013 USA) and Michael Bender (Thomson Reuters Global Resources<\/a>, Switzerland)<\/p>\n

The authors introduced a news retrieval system, eventNews<\/em>, which employs an event-centric<\/em> algorithm; thus, allowing users to monitor developing stories based on events, rather than by examining an exhaustive list of retrieved documents.<\/p>\n

News articles are clustered around an editorially supplied topical label, called a \u201cslugline\u201d. Decisions about merging related documents or clusters are made according to two distinct sources: a digital signature based on the unstructured text in the document, and the presence of named-entity tags assigned by the Thomson Reuters\u2019 Calais engine<\/a>, a named entity tagger. Human assessments were used to evaluate the system on a 5-point scale, and the average quality achieved was around 80%.<\/p>\n

The development of a more robust working model for eventNews is anticipated in the near future, while further work will focus on testing the recall of the system\u2014i.e., how many events are captured and represented from all the possible news events in the dataset or sample.<\/p>\n

Cross-Lingual Trends Detection for Named Entities in News Texts with Dynamic Neural Embedding Models<\/h5>\n

By Andrey Kutuzov (University of Oslo<\/a> \u2013 Norway) and Elizaveta Kuzmenko (National Research University Higher School of Economics<\/a> \u2013 Moscow, Russia)<\/p>\n

This paper discusses the use of vector space models, particularly \u201cneural embeddings\u201d\u2014prediction-based distributional models\u2014to detect real-world events as manifested in news texts. Temporal shifts of the embeddings might potentially predict specific events.<\/p>\n

The models are trained on a large corpus consisting of English and Russian news: English text is derived from the Signal Media dataset<\/a>, and the Russian text comes from a corpus of news articles in Russian published in September 2015 that include 500,000 extracts from about 1,000 Russian-language news websites\u2014unfortunately, not available publicly due to copyright restrictions.<\/p>\n

After training the models on the \u2018reference\u2019 corpus, they are successively updated with new textual data from daily news. The approach effectively retrieves meaningful temporal trends for named entities regardless of language. Plans to continue this work by experimenting with different algorithms or parameter sets for different languages are already on their way, and preliminary tests show promising results.<\/p>\n

Using News Articles for Real-time Cross-Lingual Event Detection and Filtering<\/h5>\n

By Gregor Leban, Bla\u017e Fortuna and Marko Grobelnik (Jozef Stefan Institute<\/a> \u2013 Ljubljana, Slovenia).<\/p>\n

This presentation referred to a system called Event Registry<\/em> which is able to group articles about an event across different languages, and extract core event information from them in a structured manner.<\/p>\n

The immediate advantage that Event Registry offers to news readers and analysts is a significant reduction on the amount of content that has to be reviewed while gathering the global coverage of a particular event. Moreover, since all the event information is structured, Event Registry provides several options for searching and filtering that are not available on existing news aggregators.<\/p>\n

According to the presenter, Gregor Leban, traditional news aggregators overwhelm users with duplicate news articles\u2014articles referring to the same event\u2014whereas the approach followed by Event Registry, based on semantic annotation per document, event clustering and the extraction of main event facts, is capable of showing events rather than articles, providing a better alternative and a more general understanding of particular events.<\/p>\n

Exploring a Large News Collection Using Visualisation Tools<\/h5>\n

By Tiago Devezas (INESC TEC<\/a> and DEI<\/a> \u2013 Porto, Portugal), Jos\u00e9 Devezas (DEI<\/a> <\/strong> <\/em> \u2013 Porto, Portugal) and S\u00e9rgio Nunes (INESC TEC<\/a> <\/strong> <\/em> and DEI<\/a> <\/strong> <\/em> \u2013 Porto, Portugal).<\/p>\n

The final paper of the session explored the Signal Media dataset<\/a> using the visualisation tools provided by the MediaViz<\/a> platform. MediaViz aims to assist in gaining insight from large archives of news through interactive visualisation tools.<\/p>\n

The visualisation analysis of the Signal Media dataset<\/a> <\/strong> <\/em> revealed the following:<\/p>\n

* News and blog sources evaluate differently the importance of similar events, granting them distinct amounts of coverage.
\n* There are both dissimilarities and overlaps in the publication patterns of the two source types\u2014news and blog sources.
\n* The content direction and diversity behave differently over time.<\/p>\n

More precisely, a keyword<\/em> analysis allowed the researchers to see that news and blog sources granted different levels of importance to a given set of keywords related with major global events that took place in September 2015. A source<\/em> analysis showed that the temporal publication patterns of these two media behaved differently\u2014blogs published a higher percentage of content during the weekend than news sources\u2014though both sources followed an identical curve during a 24-hour cycle. Finally, a diversity<\/em> analysis indicated variations in the dynamics of topical diversity over time.<\/p>\n

5\u00a0\u00a0 Poster Session<\/h2>\n

The poster session took place after finalising the second session. The accepted posters are listed below,<\/p>\n

\"\"
NewsIR'16 Poster Session<\/figcaption><\/figure>\n