Although it is under-studied relative to other social media platforms, YouTube arguably the largest and most engaging online consumption platform in world. Recently, YouTube's scale has fueled concerns that users are being radicalized via a combination of biased recommendations ostensibly apolitical "anti-woke" channels, both which have been claimed direct attention radical political content. Here we test this hypothesis using representative panel more than 300,000 Americans their...

Partisan segregation within the news audience buffers many Americans from countervailing political views, posing a risk to democracy. Empirical studies of online media ecosystem suggest that only small minority Americans, driven by mix demand and algorithms, are siloed according their ideology. However, such research omits comparatively larger television often ignores temporal dynamics underlying consumption. By analyzing billions browsing viewing events between 2016 2019, with novel...

In recent years, critics of online platforms have raised concerns about the ability recommendation algorithms to amplify problematic content, with potentially radicalizing consequences. However, attempts evaluate effect recommenders suffered from a lack appropriate counterfactuals—what user would viewed in absence algorithmic recommendations—and hence cannot disentangle effects algorithm user’s intentions. Here we propose method that call “counterfactual bots” causally estimate role...

Cyberbullying is a growing problem affecting more than half of all American teens. The main goal this paper to investigate fundamentally new approaches understand and automatically detect incidents cyberbullying over images in Instagram, media-based mobile social network. To end, we have collected sample Instagram data set consisting their associated comments, designed labeling study for as well image content using human labelers at the crowd-sourced Crowdflower Web site. An analysis labeled...

As online social networks have grown in popularity, teenage users become increasingly exposed to the threats of cyberbullying. The primary goal this research paper is investigate cyberbullying behaviors Vine, a mobile based video-sharing network, and design novel approaches automatically detect instances over Vine media sessions. We first collect set video sessions use CrowdFlower, crowd-sourced website, label for cyberaggression. then perform detailed analysis behavior Vine. Based on...

Cyberbullying is a major problem affecting more than half of all American teens. Prior work has largely focused on detecting cyberbullying after the fact. In this paper, we investigate prediction incidents in Instagram, popular media-based social network. The novelty building predictor that can anticipate occurrence before they happen. Instagram network well-suited to such since there an initial posting image typically with associated text caption, followed later by comments form basis...

Cyberbullying is a major problem affecting more than half of all American teens. Prior work has largely focused on detecting cyberbullying after the fact. In this paper, we investigate prediction incidents in Instagram, popular media-based social network. The novelty building predictor that can anticipate occurrence before they happen. Instagram network well-suited to such since there an initial posting image typically with associated text caption, followed later by comments form basis...

Online platforms have banned ("deplatformed") influencers, communities, and even entire websites to reduce content deemed harmful. Deplatformed users often migrate alternative platforms, which raises concerns about the effectiveness of deplatforming. Here, we study deplatforming Parler, a fringe social media platform, between 2020 January 11 2021 February 25, in aftermath US Capitol riot. Using two large panels that capture longitudinal user-level activity across mainstream (N = 112, 705,...

This paper examines users who are common to two popular online social networks, Instagram and Ask.fm, that often used for cyberbullying. An analysis of the negativity positivity word usage in posts by these networks is performed. These results normalized comparison a sample typical both networks. We also examine posting activity user profiles consider its correlation with negativity. Within Ask.fm network, which allows anonymous posts, relationship between anonymity further explored.

Cyberbullying is a major problem affecting more than half of all American teens, and has been attributed to suicidal behavior among teens. Instagram, media-based mobile social network, one the most popular networks used for cyberbullying. In this paper, we describe development classifiers detect cyberbullying in Instagram. We identify systems issues that need be considered design detection system.

One of the most pressing problems in high schools is bullying. However, with today's online and mobile technologies, bullying moving beyond schoolyards via cell phones, social networks, text, video images, etc. As bad as fighting were before advent Internet, recording posting hurtful content has magnified harmful reach bullying, enabling 24/7 Although cyberbullying may not cause any physical damage initially, it potentially devastating psychological effects like depression, low self-esteem,...

Cyberbullying in online social networks has become a critical problem, especially among teenagers who are networks' prolific users. As result, researchers have focused on identifying distinguishing features of cyberbullying and developing techniques to automatically detect incidents. While this research resulted highly accurate classifiers, two key practical issues related largely been ignored, namely scalability detection services timeliness raising alerts whenever incident is suspected....

In recent years, critics of online platforms have raised concerns about the ability recommendation algorithms to amplify problematic content, with potentially radicalizing consequences. However, attempts evaluate effect recommenders suffered from a lack appropriate counterfactuals -- what user would viewed in absence algorithmic recommendations and hence cannot disentangle effects algorithm user's intentions. Here we propose method that call ``counterfactual bots'' causally estimate role on...

The potential for a large, diverse population to coexist peacefully is thought depend on the existence of ``shared reality:'' public sphere in which participants are exposed similar facts about topics. A generation ago, broadcast television news was widely considered serve this function; however, since rise cable 1990s, critics and scholars have worried that corresponding fragmentation segregation audiences along partisan lines has caused shared reality be lost. Here we examine concern using...

Although it is under-studied relative to other social media platforms, YouTube arguably the largest and most engaging online consumption platform in world. Recently, YouTube's scale has fueled concerns that users are being radicalized via a combination of biased recommendations ostensibly apolitical anti-woke channels, both which have been claimed direct attention radical political content. Here we test this hypothesis using representative panel more than 300,000 Americans their...