WORKSHOP ON SOCIAL ISSUES IN ONLINE SOCIAL NETWORKS
Monday, July 23 2018
(in the context of the 2018 ACM Symposium on Principles of Distributed Computing (PODC), an international forum on the theory, design, analysis, implementation and application of distributed systems and networks)
Social issues in Online Social Networks have become part of the general interest, in part due to the political climate of the world. For example, the term fake news had a sudden spike in 2016, according to Google Trends. Efforts are being made by social networks and mass media channels to detect and prevent the spreading of fake news in social networks.
In the last couple of years we have also observed a growing interest in the auditing of online advertising in social networks and media. In this regard companies want to protect their brand by making sure their ads are not placed on websites, Facebook pages or YouTube videos whose content opposes their views, or which could associate their brand with inappropriate content. The lack of transparency in the advertisement platforms obfuscates the exact placement of ads to advertisers and the tools provided are not sufficient for the advertiser to control their brand placement.
Another important problem is that of fraud identification. Bots are being used to generate ad views or other actions that generate revenue for the website but increase the cost for the advertiser. In the US alone this kind of fraud costs $8 billion to advertisers. On the other hand, the user is concerned about how their information is being used in targeted advertisement. More transparency and privacy protection are needed. For example, in Europe there are legislations against targeted advertising based on sensitive topics such as health, politics or sexuality but we have observed that online advertising platforms are targeting users based on these topics.
In this workshop we will present methodologies, tools and techniques to audit advertising platforms addressing the issues mentioned above, as well as scalable fraud detection methods. Regarding fake news, we will present a study of three social networks: Twitter, Reddit and 4chan. We will analyze how certain communities in these networks share mainstream media and alternative media links and how communities influence each other in the same network or across networks. Furthermore, we will explore applications such as the use of social network analysis to improve Emergency Response, gender divide in social networks and the bias in mass media based on trending topics in social networks. Machine Learning and Network Science approaches to online social networking will also be considered in the workshop. One subject is transfer (or multi-task) learning, where user preference patterns extracted from certain tasks and domains can be transferred to new tasks where user data is scarce. Finally, we observe that a social network can be seen as a multilayer system, where different networks (layers) represent different relationships among its users. For example, one layer could be a friendship network, another layer could be followers or a layer of likes between users. State-of-the-art research on multilayer complex network analysis will also be presented from the point of view of network science, illustrating how both structural and dynamical properties of one network layer can be inferred by observing other network layers.
The workshop will favor interaction and discussion among participants, in order to find interesting lines of collaboration and community building
- 09:00 – 09:20 Workshop opening
- 09:20 – 10:00 Economic, cultural, and gender biases in social media – Angel Cuevas
- 10:00 – 10:40 Scalable outlier detection in online social networks – Antonio Fernandez Anta
- 10:40 – 11:10 COFFEE BREAK
- 11:10 – 11:50 Fake news propagation across online social networking platforms – Savvas Zannettou
- 12:00 – 14:00 LUNCH
- 14:00 – 14:40 Transfer learning across networks – Sotirios Chatzis
- 14:40 – 15:20 Using semantic technologies for social data sensemaking – Paloma Diaz
- 15:20 – 15:50 COFFEE BREAK
- 15:50 – 16:30 Multilayer complex network analysis and inference – Fragkiskos Papadopoulos
- 16:30 – 17:10 Auditing and behavior of online advertising – Antonio Pastor Vallés
- 17:10 – 17:30 Wrapping up discussion
Fragkiskos Papadopoulos [info]
Dept. of Electrical Engineering, Computer Engineering and Informatics, Cyprus University of Technology
30 Archbishop Kyprianou Str.
3036 Lemesos, Cyprus
Antonio Fernandez Anta [info]
IMDEA Networks Institute
Avenida del Mar Mediterraneo, 22
28918 Leganés, Madrid, Spain
Paloma Díaz [info]
Dept. of Computer Science and Engineering, Universidad Carlos III de Madrid
Avda. Universidad, 30
28911 Leganés, Madrid, Spain