News recommender systems: a programmatic research review

Abstract

News recommender systems (NRS) are becoming a ubiquitous part of the digital media landscape. Particularly in the realm of political news, the adoption of NRS can significantly impact journalistic distribution, in turn affecting journalistic work practices and news consumption. Thus, NRS touch both the supply and demand of political news. In recent years, there has been a strong increase in research on NRS. Yet, the field remains dispersed across supply and demand research perspectives. Therefore, the contribution of this programmatic research review is threefold. First, we conduct a scoping study to review scholarly work on the journalistic supply and user demand sides. Second, we identify underexplored areas. Finally, we advance five recommendations for future research from a political communication perspective.

The effects of social media usage on vicarious traumatization and the mediation role of recommendation systems usage and peer communication in China after the aircraft flight accident

“…Previous studies have found that many users keep the recommendation feature enabled on social media platforms (Nechushtai & Lewis, 2019 ). Therefore, recommendation systems can continuously affect users’ information exposure on social media (Mitova et al, 2022 ). In reality, recommendation systems frequently push information to users based on current trending topics and their browsing history (Yang, 2016 ), which may contribute to the occurrence of vicarious traumatization (Feinstein et al, 2014 ).…”

How Does Empowering Users with Greater System Control Affect News Filter Bubbles?

“…Building on pioneering research on news recommender systems by Chesnais, Mucklo, and Sheena (1995), Kamba, Bharat, and Albers (1995), and Claypool et al (1999), Mitova et al (2023) provide a systematic survey and investigation of news recommender systems in terms of how they affect journalists/media outlets (delivery perspective) and news readers (acquisition perspective). A central challenge posed by recommender systems is the lack of diversity in the recommended items (Kunaver and Požrl 2017).…”

“…We expect the interaction and transparency tool to enable "challenge-averse" individuals in the treatment group to steer the system to greater extremes than those in the control group, and enable those who are "diversity-seeking" in the treatment to steer it to less extreme and more diverse articles than those in the control. We expect mixed results for the up-vote ratio as well: the system should be able to learn user preferences better over time and hence lead to higher upvote ratios; however, some users might use the transparency and interaction tool to drastically change the system and thus experience a lower up-vote ratio than those in the control group, as providing users with more control does not guarantee its effective utilization (Mitova et al 2023).…”

“Or they could just not use it?”: The Paradox of AI Disclosure for Audience Trust in News

“…Audience Attitudes about Algorithmic Technologies and AI Despite concerns regarding audience attitudes towards AI in news production, there remains a lack of concrete evidence on the subject. User attitudes towards algorithmic technologies have so far predominantly been investigated within the domain of news recommendations (Mitova et al, 2023). As Mitova et al aruge, the theoretical foundation for much of this work is the concept of the "machine heuristic" (Sundar & Kim, 2019) which posits that algorithms are perceived positively by many due to the belief that they exhibit greater neutrality and fairness compared to humans (Mitova et al, 2023, p. 92) due to their lack of human motives and emotions.…”