Perspective-based search: a new paradigm for bursting the information bubble

TL;DR: This paper propose an adversarial nerual netowrk that decomposes a pretrained representation as independent denotation and connotation representations and show that the denotation vectors improve the viewpoint diversity of document rankings.

TL;DR: This study reports on a lab study in which participants gathered information to meet a specific type of LO, and leveraged Anderson and Krathwohl’s (A&K) taxonomy to characterize LOs and pathways.

TL;DR: A game model of bilateral evolution between information platforms and network users and simulates the influence path of key factors on the evolution of both parties’ main strategies is constructed and shows that algorithm recommendation technology is the root of “information cocoons" in the algorithm era.

TL;DR: A scientific quantitative, and visual analysis was conducted on 2101 articles in the field of information cocoons research from 1966 to 2023, finding that Twitter and Facebook are the most studied social media.

TL;DR: In this paper, the authors suggest that software designers should create tools that encourage and facilitate consumption of diverse news streams, making users, and society, better off, and propose several techniques to help achieve this goal.

TL;DR: Experimental results strongly support that ideological perspectives are reflected in lexical variations in the proposed statistical model for ideology discourse.

TL;DR: There is concrete evidence for a tendency that users predominantly share like-minded news articles and avoid conflicting ones, and partisans are more likely to do that in social media.

TL;DR: This research broadens previous work on biases in search to examine the role of search systems in contributing to biases, and focuses on questions about medical interventions and employs reliable ground truth data from authoritative medical sources to assess bias.