aisa6148
commented
1 week ago In an experimental study using a custom-built news app, we show that diversity-optimized recommendations
- perform similar to methods optimizing for user preferences regarding user utility
- that diverse news recommendations are related to a higher tolerance for opposing views, especially for politically conservative users
- that diverse news recommender systems may nudge users towards preferring news with differing or even opposing views. We conclude that diverse news recommendations can have a depolarizing capacity for democratic societies.
RQ 1.: How do diverse news recommendations affect news consumption? RQ 2.: How do diverse news recommendations affect individual utility of news consumption? RQ 3.: How do diverse news recommendations affect societal externalities of news consumption?
- news recommender systems “should not just maximize for clicks and short-term revenue, but, mindful of the democratic function of the media, also optimize for values that align with the overall mission of a news outlet”
- low exposure leads to low voter turn out
- Diversity is a design design to enhance positive externalities of NRS
- Participants split into diversity (n=28), narrow = (n=35) and control i.e. chronologically ranked news (n=88)
- Diversity in this app: defining diversity as a discretized distribution of news articles over a political spectrum.
- hybrid, human-centered scoring pipeline for news articles that is based on the political position of users.
- Article political alignment is determined by aggregated average scores of the readers' political alignment. It does not actually rate the article, rather labels the political stance of its readers based on implicit and explicit. this is claiming to reduce bias from experts or content filtering
- User scoring: 20+ questions for the users -> each question 100points -> multiplied by the score they gave (0.74 for 4, 0.5 for 3, 0.25 for 2, 0 for 1) -> weighted of 1 or -1 added to the questions -> normalized between -1 and 1
- Article scoring
- different political stance and average
- Perceived diversity measured: “the app covered a broad spectrum of news,” “was impartial regarding political opinions,” “covered multiple political opinions,” and “fit to my own political opinion.” - results as expected with each group
- Instrumental utility: “news in the app helps me to understand our society,” “helps me to make wise decisions,” “provides me with a daily account of what is happening in the world” - Result: Not as significant/No relation
- Social Externalities: (a) political factors (such as knowledgeability and participation) -- result: moderate (b) tolerance of opposing views, -- result: especially for conservatives (c) social performance of journalism, and (d) news diversity preferences. Regarding political factors, we measure political knowledge, interest, and efficacy with one item each
- participants moderately believe that news has the ability to help society
- conservatives agree more to journalism's capacity
- Tolerance to opposing views was higher in diversity group
- results suggest that recommender systems may nudge users to preferences according to a prevalent recommender design. While heavy news users across all groups prefer opposing views, the diversity condition may nudge all its users to appreciate cross-cut- ting content.
- UTILITY IS A STRONG DETERMINANT
- The participants were not not aware of the conditions
- Diversity my have de-polarizing capacity
- Future: The current rating and recommendation pipeline ignores the content of news articles. It solely focuses on reading metrics and ratings of the participants. Hence, the political labels of articles might say more about their readership than about the article contents. In order to account for this shortcoming, entity and sentiment recognition would need to be implemented.
- Future: experiment with varying threshold for diversity
Digital Journalism, 10(10), 1710-1730. Heitz, L., Lischka, J. A., Birrer, A., Paudel, B., Tolmeijer, S., Laugwitz, L., & Bernstein, A. (2022)