Best Paper Award: Alexandra Olteanu, Anne-Marie Kermarrec, Karl Aberer, “Comparing the Predictive Capability of Social and Interest Affinity for Recommendations”. WISE (1) 2014: 276-292
Best Paper at WISE 2014
Oct 14th, 2014 by arault
D1 seminare : talk of Nitin Chiluka on April the 22nd in Sicile room at 03:00PM
Mar 27th, 2014 by cbouton
Title :
Securing Social Media via Network Structure
Abstract:
Due to its democratized nature, online social media (OSM) attracts millions of users to publish and share their content with friends as well as a wider audience at little cost. Such a vast user base and a wealth of content, however, presents its own challenges. First, the amount of user-generated content being uploaded in their repositories makes it non-trivial for users to find relevant content. Second, the ease of creating an account and participating in OSM enables malicious users to spread and promote spam content while degrading the experience for normal users. Third, growing privacy concerns in OSM motivate users to move toward decentralized social networks which are still in a nascent stage.
In our work, we propose a network structure approach to address these challenges in OSM. We build scalable and effective graph-based algorithms that recommend content personalized to each user. To tackle adversarial environments, we leverage the social network graph as well as negative feedback from normal users to limit the capability of knowledgeable attackers in multiple scenarios. Further, we highlight some of the practical limitations in existing efforts to build decentralized social networks, and outline desiderata for such distributed systems to work well in the wild.
Forum Liberation Rennes 2014
Mar 21st, 2014 by akermarrec
Talk of Vinay Setty from the University of Oslo, Norway on Friday, February 28th at 10:00am in Lipari room.
Feb 25th, 2014 by cbouton
Title :
Publish/Subscribe for Large-Scale Social Interaction: Design, Analysis and Resource Provisioning
Abstract:
Publish/subscribe (pub/sub) is a popular communication paradigm in the design of large-scale distributed systems. We are witnessing an increasingly widespread use of the pub/sub for wide array of applications both in industry and academia and yet there is a lack of detailed study of a large-scale real-world pub/sub system. In our work we present an overview of a pub/sub system used to drive social interaction at Spotify. We then present a detailed analysis of the traces from real deployment of Spotify pub/sub. Inspired by the peer-assisted solution used by Spotify to stream music, we explore a similar solution to disseminate messages of Spotify pub/sub to the users. The task of distributing the workload among user peers and datacenter servers prompts a fundamental problem: How to select a subset of pub/sub workload to be served by datacenter servers in a manner to maximise satisfaction requirements of users under resource constraints?
In our recent work, we provide, to the best of our knowledge, the first formal treatment of the above problem by introducing two metrics that capture subscriber satisfaction in the presence of limited resources. This allows us to formulate the problem as two new flavors of maximum coverage optimization problems. Unfortunately, both variants of the problem prove to be NP-hard. By subsequently providing formal approximation bounds and heuristics, we show, however, that efficient approximations can be attained. We validate our approach using real-world traces from Spotify and show that our solutions can be executed periodically in real-time in order to adapt to workload
variations.
Further, we try to answers to the following three fundamental questions: Given a pub/sub workload, (1) what is the minimum amount of resources needed to satisfy all the subscribers, (2) what is a cost-effective way to allocate resources for the given workload, and (3) what is the cost of hosting it on a public Infrastructure-as-a-Service (IaaS) provider like Amazon EC2.