A Collusion-Resistant Automation Scheme for Social Moderation Systems

Abstract

Abstract-For current Web 2.0 services, manual examination of user uploaded content is normally required to ensure its legitimacy and appropriateness, which is a substantial burden to service providers. To reduce labor costs and the delays caused by content censoring, social moderation has been proposed as a front-line mechanism, whereby user moderators are encouraged to examine content before system moderation is required. Given the immerse amount of new content added to the Web each day, there is a need for automation schemes to facilitate rear system moderation. This kind of mechanism is expected to automatically summarize reports from user moderators and ban misbehaving users or remove inappropriate content whenever possible. However, the accuracy of such schemes may be reduced by collusion attacks, where some work together to mislead the automatic summarization in order to obtain shared benefits.In this paper, we propose a collusion-resistant automation scheme for social moderation systems. Because some user moderators may collude and dishonestly claim that a user misbehaves, our scheme detects whether an accusation from a user moderator is fair or malicious based on the structure of mutual accusations of all users in the system. Through simulations we show that collusion attacks are likely to succeed if an intuitive countbased automation scheme is used. The proposed scheme, which is based on the community structure of the user accusation graph, achieves a decent performance in most scenarios.

d.tour

“…People can already add pages to d.tour's database with the search box; given enough users, the database could expand by crowdsourcing. Social moderation techniques [20] could help maintain quality as the corpus grows organically.…”

Soft Play Detection in Shooter Games Using Hit Matrix Analysis

“…Depending on the algorithm this can be the largest or the smallest cluster, but using such a simple criterion will miss some of the correct subsets. Options for selecting the colluding subset can be based in-degree/out-degree [12] or the collusion index Φ(C) = I(C)/E(C), where I(C) is the sum of edges where both vertices are in C, and E(C) is the sum of edges where only other vertex is in C [10]. When using the plain hit matrix values to generate a graph, the inverse of the collusion index is used.…”

“…Van der Knyff et al [5] try to find the colluding subset in a first-person-shooter game using principal component analysis [6]. Outside the field of multiplayer games, collusion has been addressed in tournaments [7], multiple choice examinations [8], covert communication channels [9], stock market trading [10], grid computing [11], social moderation [12] and spectrum auctions [13].…”

Detecting false testimonies in reputation systems using self-organizing maps

“…A typical solution is the one given in [5] that relies on cryptography. However, with the existence of side channel attacks [6], the attacker can easily guess the secret keys and compromise the cryptography-based protocols.…”

“…However, this imposes restrictions on training data, as if the algorithm does not process the 'unclean' data during the training, it will not be able to detect attacks. A solution based on graph theory is given in [9]. This solution, instead of using the countbased scheme that considers the number of accusations, uses the community-based scheme that achieves the detection of up to 90% of the attackers, which permits the correct operation of the system.…”