3bij3 – Developing a framework for researching recommender systems and their effects

Löecherbach, Felicia; Trilling, Damian

doi:10.5117/ccr2020.1.003.loec

Cited by 14 publications

(16 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike much passive "television-era" theorizing, consumers and media companies are now in a kind of interplay. Beyond media "cultivation," this new interplay could be thought of as a kind of mutual cultivation: Producers and consumers now exhibit much more co-construction of the social reality through real-time engagement and experimentation in Internet and social media (Beer, 2009;Loecherbach & Trilling, 2020;Rieder, Matamoros-Fernández, & Coromina, 2018;Varian, 2014;Valkenburg et al, 2016). Some researchers have sought to revise and extend traditional theories in this new context, asserting their continued relevance (Morgan et al, 2015).…”

Section: Discussion and Theoretical Implicationsmentioning

confidence: 99%

The CogMedia Project: Open Data and Tools for Linking Cognitive Science and Mass Media

Dale¹

2020

Preprint

View full text Add to dashboard Cite

Minds produce and consume news media. Because of this central relevance of minds, theories of mass media may be informed by the methods and measures of cognitive science. These measures could enhance understanding of the factors guiding news consumption habits. These measures could also help to explain how news themes accumulate collective attention. CogMedia is introduced here as an open resource for facilitating a connection between mind and mass communication. It contains hundreds of thousands of major headlines from large news media organizations, primarily in the United States. In addition, it has a set of open analysis tools for conducting quantitative analysis in R. This paper motivates CogMedia with background theoretical and empirical review. Theoretical discussion centers on complexity science as a framework for understanding mass-scale systems and the cognitive agents that underlie them. Following this background review and motivation, the paper introduces CogMedia, and illustrates potential quantitative analysis and data visualization using these open tools. Future empirical work is discussed, along with relevant theoretical implications.

show abstract

Section: Discussion and Theoretical Implicationsmentioning

confidence: 99%

The CogMedia Project: Open Data and Tools for Linking Cognitive Science and Mass Media

Dale¹

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In the reality of news companies, it is challenging to measure satisfaction with the recommendation beyond the actual click (Bodó, 2019) as additional user surveys are required. In communication research, it is in turn dif-ficult to simulate realistic article recommendation and study recommendations as to the interaction between algorithms and the user (Loecherbach & Trilling, 2020). Data on user views of real recommendations are accordingly scarce in academia: most are derived from hypothetical instructions.…”

Section: Methodsmentioning

confidence: 99%

One Recommender Fits All? An Exploration of User Satisfaction With Text-Based News Recommender Systems

Wieland¹,

Nordheim²,

Königslöw³

2021

MaC

View full text Add to dashboard Cite

Journalistic media increasingly address changing user behaviour online by implementing algorithmic recommendations on their pages. While social media extensively rely on user data for personalized recommendations, journalistic media may choose to aim to improve the user experience based on textual features such as thematic similarity. From a societal viewpoint, these recommendations should be as diverse as possible. Users, however, tend to prefer recommendations that enable “serendipity”—the perception of an item as a welcome surprise that strikes just the right balance between more similarly useful but still novel content. By conducting a representative online survey with n = 588 respondents, we investigate how users evaluate algorithmic news recommendations (recommendation satisfaction, as well as perceived novelty and unexpectedness) based on different similarity settings and how individual dispositions (news interest, civic information norm, need for cognitive closure, etc.) may affect these evaluations. The core piece of our survey is a self-programmed recommendation system that accesses a database of vectorized news articles. Respondents search for a personally relevant keyword and select a suitable article, after which another article is recommended automatically, at random, using one of three similarity settings. Our findings show that users prefer recommendations of the most similar articles, which are at the same time perceived as novel, but not necessarily unexpected. However, user evaluations will differ depending on personal characteristics such as formal education, the civic information norm, and the need for cognitive closure.

show abstract

“…To answer the research questions, a platform for testing an interactive news recommender system for a longer period of time is used [16]. It is based on an open-source Python application and presents the user with a web interface showing a selection of nine different news articles in tiles arranged in a 3x3 grid.…”

Section: Methodsmentioning

confidence: 99%

“…For each article, the title and a short teaser are shown, additionally the topic is indicated with a colored tag. For more information on the interface and setup of the website, a detailed description can be found in [16]. The news are retrieved from several RSS feeds of different Dutch news providers (similar to strategies employed in [24]).…”

Section: Methodsmentioning

confidence: 99%

“…The last group was in charge of the personalization by having the option to state and adapt their topic preferences (customization: n = 42). For more detailed information on the customization and recommendation algorithms used see [16]. For all groups but the random group 6 out of the 9 stories were personalized, either according to their explicit wishes (customization) or based on their behavior (recommendation algorithm).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Is this a click towards diversity? Explaining when and why news users make diverse choices

Löecherbach

Welbers

Moeller

et al. 2021

13th ACM Web Science Conference 2021

Self Cite

View full text Add to dashboard Cite

Modelling the different factors that lead people to choose news articles is one of the key challenges for understanding the diversity of news diets -as a news diet is the result of a series of decisions for certain articles over others, a sequence of choices that was made by the individual consumer. This study sheds light on the interplay between content-related (past behavior, habits, preferences) and situational factors (positioning, saturation, control). The latter could offer possibilities to promote more unexpected news encounters that diverge from past news consumption. To test this, a Pythonbased web application for interactively testing different forms of news personalization over time was used. 247 respondents used the system over a two-week period, in total making almost 23,000 choices. Our results show that: (1) Selections are influenced by a strong positioning effect that follows a reading pattern (left-right, up-down). This effect is stable across devices, topics, and preferences. (2) How much control people are given influences the length and the amount of different sessions (personalization leads to fewer and shorter sessions). (3) With high control, the diversity of preferences influenced the diversity of selected news more, possibly widening gaps between diversity-seeking and -averse users. (4) How often a topic was chosen in the last hour negatively impacts whether it gets chosen again, showing saturation effects. (5) Clicks on sports and economic articles can be explained by preferences, but not past behavior; for political news the opposite is found. (6) There is no significant correlation between the actual diversity (presented or selected topics) and the topic diversity perceived by the users -in spite of clear differences in actual diversity between the groups. From this we can conclude the importance of situational factors in modelling news selection and their potential to narrow This work is licensed under a Creative Commons Attribution International 4.0 License.

show abstract

3bij3 – Developing a framework for researching recommender systems and their effects

Cited by 14 publications

References 0 publications

The CogMedia Project: Open Data and Tools for Linking Cognitive Science and Mass Media

The CogMedia Project: Open Data and Tools for Linking Cognitive Science and Mass Media

One Recommender Fits All? An Exploration of User Satisfaction With Text-Based News Recommender Systems

Is this a click towards diversity? Explaining when and why news users make diverse choices

Contact Info

Product

Resources

About