The Interaction between Political Typology and Filter Bubbles in News Recommendation Algorithms
The Interaction between Political Typology and Filter Bubbles in News Recommendation Algorithms
复制标题
DOI：
10.1145/3442381.3450113
复制
发表时间：
2021-04
期刊：
影响因子：
--
通讯作者：
Ping Liu;K. Shivaram;A. Culotta;Matthew A. Shapiro;M. Bilgic
中科院分区：
文献类型：
--
作者：
Ping Liu;K. Shivaram;A. Culotta;Matthew A. Shapiro;M. Bilgic
新闻和社交媒体内容的个性化旨在改善用户体验;然而，有证据表明，这种过滤可能会产生意想不到的副作用，即产生同质的“过滤气泡”，用户过度接触与他们先前存在的感知和信念相符的想法。在本文中，我们调查这一现象的背景下，政治新闻推荐算法，这对公民话语具有重要意义。我们首先从41个来源收集和整理了超过90万篇新闻文章，并按主题和党派倾向进行了注释。然后，我们进行模拟研究，探讨不同的算法策略如何影响过滤器气泡的形成。借鉴皮尤研究的政治类型学，我们确定基于用户的预先存在的偏好异质性的影响。例如，我们发现，i）具有更极端偏好的用户显示的内容不太多样化，但比具有不那么极端偏好的用户具有更高的点击率，ii）基于内容和协作过滤的过滤器导致明显不同的过滤气泡，以及iii）当用户对不同主题有不同的观点时，过滤器往往具有同质化效应。
Algorithmic personalization of news and social media content aims to improve user experience; however, there is evidence that this filtering can have the unintended side effect of creating homogeneous “filter bubbles,” in which users are over-exposed to ideas that conform with their preexisting perceptions and beliefs. In this paper, we investigate this phenomenon in the context of political news recommendation algorithms, which have important implications for civil discourse. We first collect and curate a collection of over 900K news articles from 41 sources annotated by topic and partisan lean. We then conduct simulation studies to investigate how different algorithmic strategies affect filter bubble formation. Drawing on Pew studies of political typologies, we identify heterogeneous effects based on the user’s pre-existing preferences. For example, we find that i) users with more extreme preferences are shown less diverse content but have higher click-through rates than users with less extreme preferences, ii) content-based and collaborative-filtering recommenders result in markedly different filter bubbles, and iii) when users have divergent views on different topics, recommenders tend to have a homogenization effect.