Helping Users Tackle Algorithmic Threats on Social Media: A Multimedia Research Agenda

Weth, Christian von der; Abdul, Ashraf; Kankanhalli, Mohan

doi:10.1145/3394171.3414692

Cited by 13 publications

(3 citation statements)

References 99 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Photos constitute a large part of the data shared on OSNs. A wealth of information can be inferred from their content and used against users [48]. Existing works usually classify user images as private/public.…”

Section: Related Workmentioning

confidence: 99%

Raising User Awareness about the Consequences of Online Photo Sharing

Schindler

Popescu

Nguyen

et al. 2023

Proceedings of the 2023 ACM International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

Online social networks use AI techniques to automatically infer profiles from users' shared data. However, these inferences and their effects remain, to a large extent, opaque to the users themselves. We propose a method which raises user awareness about the potential use of their profiles in impactful situations, such as searching for a job or an accommodation. These situations illustrate usage contexts that users might not have anticipated when deciding to share their data. User photographic profiles are described by automatic object detections in profile photos, and associated object ratings in situations. Human ratings of the profiles per situation are also available for training. These data are represented as graph structures which are fed into graph neural networks in order to learn how to automatically rate them. An adaptation of the learning procedure per situation is proposed since the same profile is likely to be interpreted differently, depending on the context. Automatic profile ratings are compared to one another in order to inform individual users of their standing with respect to others. Our method is evaluated on a public dataset, and consistently outperforms competitive baselines. An ablation study gives insights about the role of its main components.

show abstract

Section: Related Workmentioning

confidence: 99%

Raising User Awareness about the Consequences of Online Photo Sharing

Schindler

Popescu

Nguyen

et al. 2023

Proceedings of the 2023 ACM International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

show abstract

“…With millions of daily active users, Twitter is a thriving social media platform with significant influence in shaping public opinion. However, as well as bringing benefits, social media also poses a major threat [37]. The platform's immense power has led to the proliferation of a new type of users, known as social bots.…”

Section: Introductionmentioning

confidence: 99%

My Brother Helps Me: Node Injection Based Adversarial Attack on Social Bot Detection

Wang,

Qiao,

Xie

et al. 2023

Proceedings of the 31st ACM International Conference on Multimedia

View full text Add to dashboard Cite

Social platforms such as Twitter are under siege from a multitude of fraudulent users. In response, social bot detection tasks have been developed to identify such fake users. Due to the structure of social networks, the majority of methods are based on the graph neural network(GNN), which is susceptible to attacks. In this study, we propose a node injection-based adversarial attack method designed to deceive bot detection models. Notably, neither the target bot nor the newly injected bot can be detected when a new bot is added around the target bot. This attack operates in a black-box fashion, implying that any information related to the victim model remains unknown. To our knowledge, this is the first study exploring the resilience of bot detection through graph node injection. Furthermore, we develop an attribute recovery module to revert the injected node embedding from the graph embedding space back to the original feature space, enabling the adversary to manipulate node perturbation effectively. We conduct adversarial attacks on four commonly used GNN structures for bot detection on two widely used datasets: Cresci-2015 and TwiBot-22. The attack success rate is over 73% and the rate of newly injected nodes being detected as bots is below 13% on these two datasets. CCS CONCEPTS• Computing methodologies → Machine learning algorithms; • Security and privacy → Human and societal aspects of security and privacy.

show abstract

“…In this context, intelligent, machine learning-based algorithms appear in different roles: On the one hand, they are part of the problem, e.g., when personalized targeting of information supports the creation of filter bubbles. On the other hand, they can help to detect threats and generate supportive scaffolds to counter such risks (von der Weth et al, 2020). The detection of hate speech (MacAvaney et al, 2019) is of particular interest for providers of social media platforms.…”

Section: Introductionmentioning

confidence: 99%

How to measure disagreement as a premise for learning from controversy in a social media context

Malzahn¹,

Schwarze²,

Eimler³

et al. 2022

RPTEL

View full text Add to dashboard Cite

Learning scenarios building on disagreement in a learning group or a whole classroom are well established in modern pedagogy. In the specific tradition of collaborative learning, such approaches have been traced back to theories of socio-cognitive conflict and have been associated with argumentative learning interactions. An important premise for these types of learning scenarios is the identification of disagreement. In the spirit of learning analytics, this calls for analytic tools and mechanisms to detect and measure disagreement in learning groups.Our mathematical analysis of several methods shows that methods of different origin are largely equivalent, only differing in the normalization factors and ensuing scaling properties. We have selected a measure that scales best and applied it to a target scenario in which learners judged types and levels of “toxicity” of social media content using an interactive tagging tool. Due restrictions imposed by the pandemic, we had to replace the originally envisaged classroom scenario by online experiments. We report on two consecutive experiments involving 42 students in the first and 89 subjects in the second instance. The results corroborate the adequacy of the measure in combination with the interactive, game-based approach to collecting judgements. We also saw that a revision of categories after the first study reduced the ambiguity. In addition to applying the disagreement measure to the learner judgements, we also assessed several personality traits, such as authoritarianism and social closeness. Regarding the dependency of the learner judgements on personality traits, we could only observe a weak influence of authoritarianism.

show abstract

Helping Users Tackle Algorithmic Threats on Social Media: A Multimedia Research Agenda

Cited by 13 publications

References 99 publications

Raising User Awareness about the Consequences of Online Photo Sharing

Raising User Awareness about the Consequences of Online Photo Sharing

My Brother Helps Me: Node Injection Based Adversarial Attack on Social Bot Detection

How to measure disagreement as a premise for learning from controversy in a social media context

Contact Info

Product

Resources

About