First place for CCS.Amsterdam members in Dutch Public Broadcaster’s Hackathon On Diverse Recommenders!

On 9 and 10 October, a UvA-VU team consisting of Myrthe Reuver, Felicia Loecherbach, Nicolas Mattis, Sanne Vrijenhoek, and Marijn Sax took part in a 24-hour hackathon organized by the Dutch Public Broadcaster NPO. The teams were ask to help build solutions to promote diverse content to NPOstart users. The Geeky Griffins (as the team was called) built a new explore feature, as well as a match me up feature which — in line with a longstanding Dutch tradition — allows users to sneak a peek into the digital living room of other people through their open curtains. After deliberations by an expert jury, The Geeky Griffins were declared the winners of the hackathon.

Congratulations!

More information about the hackathon can be found on the mediaperspectives website.

Marieke and Nicolas join the team!

We are happy to announce that two new and very talented PhD students will join CCS.amsterdam team in September:

Marieke van Hoof will be researching Polarizing Issue Publics: the interplay of human choice and algorithmic selection, as part of the Digital Society initiative at the Amsterdam School of Communication Research, UvA. In this we will study the role of the digital media environment and AI in causes and effects of polarizing and radicalizing issue publics using two case studies, for example, the anti-5G movement, and the vaccination movement. Marieke has already been a part of CSS Amsterdam during her research master in Social Science, and we are very happy to welcome her back. She will be supervised by Judith Möller, Corine Meppelink, and Damian Trilling.

Nicolas Mattis will work on the NWO funded project Rethinking News Algorithms, where he will use lab and field experiments to model how news users react to news recommendations: what stories do they actually read, does it matter how the stories are presented, and how can we fine-tune the recommender algorithm to nudge them to consume a more balanced news diet? Nicolas is originally from Germany and did a research master in Communication Science at the UvA. During his research master, he focused on using automated research methods to study how digitalisation and commercialisation impact journalism and society. He will be supervised by Wouter van Atteveldt and Judith Möller.

Welcome both to the team and we are very much looking forward to working with you to answer the pressing questions you will be working on!

Two new grants for ccs.amsterdam

We are starting the holiday season in a very festive mood with good news on two ccs.amsterdam grant applications related to news algorithms and diversity:

Rethinking news algorithms: nudging users towards diverse news exposure: We improve news algorithms to stimulate people to read more diverse news. Algorithms such as used by Facebook and Google can unwittingly trap people in a “filter bubble”. Nudging people to read about more topics and perspectives makes them more aware of issues facing the country.

Algorithms for freedom of expression and a well-informed public: Recent concerns about filter bubbles, fake news and selective exposure are symptomatic of the disruptive potential of digital recommendations, but also demonstrate the power of responsible recommender design Much depends on the design of news recommenders. Are they merely designed to generate clicks and short-term engagement? Or are they programmed to balance short term engagement and relevancy with the legitimate longer-term interest of users in diverse information and not losing out on important information?

3 VENI grants for ccs.amsterdam researchers

Fantastic news to start the summer: three ccs.amsterdam researchers were granted the prestigious NWO VENI individual research grants of 250kE each. Congratulations to Judith Möller, Mariken van der Velden, and Kasper Welbers!

See brief project summaries below, and check back here later to see the full project descriptions and research progress.

Vocal, Visible and Vaulting? (Dis)connected niche audiences in the age of artificial intelligence
Dr J.E. (Judith) Möller, UvA – Amsterdam School of Communication Research

This project studies the impact of algorithmic filter systems and artificial intelligence on specific segments of the population and niche audiences (fringe bubbles). It will contribute to our understanding of the consequences of AI for the public debate and the cohesion of the public sphere.

A Promise is a Promise! Unfolding the Electoral Ramifications of Political Compromises
Dr M.A.C.G. (Mariken) van der Velden, VU

Nowadays, political compromises seem to be extremely unpopular. This poses a problem for politicians’ ability to form stable coalitions to govern. My project studies the ramifications of compromises for politicians’ communication and voter behaviour utilizing virtual reality techniques, experiments and computational textual approaches.

Gatekeeping in the era of fake news: The effects of trust on online news consumption behaviour
Dr K. (Kasper) Welbers, VU – Communication Science

Despite much debate about the state of trust in online news sources, we know little about the effects of trust on online news consumption behavior. I link digital trace data with a panel-survey and content analysis to investigate these effects in unprecedented detail to enrich the media trust debate.

Computational Communication Research: Inaugural Issue

We are very happy to announce the inaugural issue for Computational Communication Research! The articles are currently in production, but you can access the preprints using the links below.  Please help us spread the word!

We would like to thank all reviewers, submitters, and editorial board members for contributing to the journal and for their feedback on this introduction. We would also like to thank Amsterdam University Press and especially our gold sponsors (Vrije Universiteit AmsterdamThe Network Institute, the University of Amsterdam / ASCoR) and silver sponsors (The Hebrew University of Jerusalem, The Center for Information Technology and Society at UC Santa Barbara, and the Computational Communication Science Lab of the University of Vienna), for making this journal possible.Looking forward to your submissions and reviews in the coming months!CCR Inaugural Issue[https://computationalcommunication.org/index.php/ccr/announcement/view/1]

A Roadmap for Computational Communication Research [draft version]
Wouter van Atteveldt, Drew Margolin, Cuihua Shen, Damian Trilling, René Weber
[https://osf.io/preprints/socarxiv/4dhfk]

iCoRe: The GDELT Interface for the Advancement of Communication ResearchFrederic R. Hopp, Jacob T. Fisher, René Weber
[https://osf.io/preprints/socarxiv/smjwb]

An Experimental Study of Recommendation Algorithms for Tailored Health Communication
Hyun Suk Kim, Sijia Yang, Minji Kim, Brett Hemenway, Lyle Ungar, Joseph Cappella:
[https://osf.io/preprints/socarxiv/nu6tg]

News Organizations’ Selective Link Sharing as Gatekeeping: A Structural Topic Model Approach
Chankyung Pak
[https://osf.io/preprints/socarxiv/pt7es]

Computational observation: Challenges and opportunities of automated observation within algorithmically curated media environments using a browser plug-in
Mario Haim, Angela Nienierza
[https://osf.io/preprints/socarxiv/xd63n/]

New project: SHARENEWS

We got a new project – SHARENEWS! We are happy to announce that we got funding (and data access) to conduct a multi-national project on the shareworthiness of news on Facebook. The multi-national project is coordinated by a PI from ccs.amsterdam, Damian Trilling; and also Judith Möller, Wouter van Atteveldt, and Claes de Vreese are part of the Amsterdam team. See here for the whole team.

CCS Amsterdam @ICA 2019

We are happy to share that a large number of our submissions for the upcoming conference of the International Communication Association (ICA 2019) were accepted.

Gathering Mobile News Consumption Traces: An Overview of Possibilities and a Prototype Tool based on Google Takeout
Wouter van Atteveldt, Laurens Bogaardt, Vincent van Hees,  Felicia Loecherbach, Judith Moeller, Damian Trilling

This paper gives an overview of technical, legal, and ethical possibilities and challenges in collecting digital trace data of mobile news consumption. The shift to online and especially mobile news consumption has led to the creation of unprecedented amounts of data on individual news consumption behavior. In theory, we can now know which respondents viewed (and liked, shared, and commented on) which news articles at which time. This allows a much more fine-grained study of media effects compared to the ‘classical’ method of linking media content to (panel) survey data. In this paper, we first discuss the criteria and desirable aspects for a possible solution. We then present an overview of different possibilities and discuss the extent to which each solution meets the listed criteria. Based on this discussion, we then present a prototype for using the activity log download or ‘takeout’ functionality offered by modern platforms to construct a similar system to the Web Historian discussed above.

It takes three to tango: The interplay of political press releases, social media and newspapers
Felicia Loecherbach, Damian Trilling

If a party wants to communicate issues, it issues a press release which gets picked up by the media and distributed to the audience. This “tango” of politics and media might have been altered over the past decade by the advent of a third partner: social media. In this study we determine whether signs of substitution of press releases by social media activity can be seen over time, indicating a change in the communication channels used by political parties. Second, we look at the influence of press releases on press coverage to determine to which extent the former exhibits influence on the latter. This sheds the light on the development of press releases’ agenda setting function and power to influence the public discourse via media outlets. Time series analysis and similarity calculations are applied to datasets including party communications (press releases, Facebook posts) and media content of the past decade.

3bij3 – A framework for testing effects of recommender systems on news exposure
Felicia Loecherbach, Damian Trilling

The impact of recommender systems on how we consume and perceive news is still understudied. To shed more light on this, we developed one of the first software solutions for such studies, and present 3bij3. It consists of a news application interface for study participants that displays a 3×3 grid of the nine most relevant news articles to a user, selected by different mechanisms (e.g., explicit customization vs based on past behavior). Compared to previous work, 3bij3 gives researchers control over the recommendation system under study and creates a realistic environment for the participants. It integrates web scraping/parsing, supervised machine learning for labeling articles, different recommender systems, a web interface for participants, and gamification elements (earning points for interaction). 3bij3 can be used to conduct large-scale field experiments, in which participants’ use of the site can be tracked over extended periods of time.

Finding your way: How news consumers seek and find news about different topics on online platforms
Susan Vermeer

The complexity and diversity of today’s media landscape provides many challenges for scholars studying online news consumption. Yet it is unclear how news consumers navigate the wealth of online news. To move forward, we used a custom-built system–passively tracking online news consumers 24/7–to examine how context (i.e., platform) and content (i.e., news topic) features affect patterns of online news consumption. This bottom-up design resulted in a large data set containing more than 1 million news-related Web pages, from 165 different news websites, as well as search engines and social media, collected over 8 months in 2017/18. We found that news consumers often directly visit the homepage of their favorite (typically mainstream) news outlet, and continue browsing within that same news outlet. Our findings also show a strong preference for entertainment news over any other topic. Social media frequently expose users to entertainment news, but are not necessarily the starting point to this type of news.

Guilty by Association: Using Word Embeddings to Measure Ethnic Stereotypes in News Coverage
Anne C. Kroon, Damian Trilling, Tamara Raats

The current study provides a new level of empirical evidence for the nature of ethnic stereotypes in news coverage by drawing on a sample of more than three million Dutch news items. The study’s findings demonstrate that universally-accepted dimensions of stereotype content (i.e., low-status and high-threat attributes) can be replicated in news media coverage across a diverse set of ingroup and outgroup categories. News representations of minorities have become progressively negative and remote from factual integration outcomes, and are therefore rather an artifact of news production processes than a true reflection of what is actually happening in society.

New member: Felicia Loecherbach

We are extremely happy to announce that we have hired Felicia Loecherbach, who will be joining us as PhD student on the Filter Bubbles Project.

Felicia combines a strong communication science background with very good technical skills. She completed the UvA research master in Communication Science. For her thesis she implemented her own news recommender system and web interface in python for conducting live, large-scale experiments tracking usage over time. It combines web scraping, machine learning, different customization options as well as gamification elements to enable testing various forms of news personalization in a controlled, yet realistic environment.

In the filter bubbles project, she will concentrate on the substantive analysis of (mobile) digital traces of news consumption behaviour: what news are people actually exposed to and choose to consume, can we find substantial evidence for echo chambers or filter bubbles, and if so what are the possible consequences for political knowledge and attitudes? Her other research interests include the interplay between political public relations and (social) media as well as furthering the usage of computational methods in communication science. She contributes for example to the open-source project INCA to facilitate collection, processing and analysis of online (textual) data for social scientists.

Since we now also know that Vincent van Hees and Laurens Bogaardt are joining us from the Netherlands eScience Center as eScience engineers, the team is complete and we can get to work 🙂

New journal: Computational Communication Research

We am very excited to announce that we just launched Computational Communication Research (CCR), a new open-access peer-reviewed journal dedicated to development and applications of computational methods for communication science. We hope that CCR will serve as central home for communication scientists with an interest in and focus on computational methods — a place to read and publish the cutting edge work in our growing subfield.

Please see the inaugural call for papers at http://computationalcommunication.org/inaugural-cfp/ (abstracts 30 Sept, manuscripts 30 Nov), and consider submitting your best computational work to the first issue!

Don’t hesitate to contact us for more information, and looking forward to your submissions!