Special Issue
Selected papers of the ACM SAC 2005 TRECK Track
Table of contents
- Foreword to the ‘ACM SAC 2005 TRECK Track’ Selected Papers
J.-M. Seigneur. abstract - How to Incorporate Revocation Status Information into the Trust Metrics for Public-Key Certification
K. Bicakci, B. Crispo, A.S. Tanenbaum. abstract - Trusting Pseudonyms: Anonymous Communication in Peer-to-Peer Reputation Systems,
M. Kinateder, R. Terdic, K. Rothermel. abstract - Object Trust Management for Information Quality Assurance in Virtual Organisations
Y. Zuo, B. Panda. abstract - Moleskiing.it: a Trust-Aware Recommender System for Ski Mountaineering
P. Avesani, P. Massa, R. Tiella. abstract
Paper abstracts
Foreword to the ‘ACM SAC 2005 TRECK Track’ Selected Papers
Jean-Marc Seigneur
Full paper (PDF, 49Kb)
How to Incorporate Revocation Status Information into the Trust Metrics for Public-Key Certification
Kemal Bicakci
Bruno Crispo
Andrew S. Tanenbaum
Full paper (PDF, 268Kb)
In order to validate a digital signature, there should be an authentic link between the corresponding public key and its owner. In a traditional PKI, the trust associated with this link is expressed in binary either by 0 or 1. Alternatively, several authors have proposed trust metrics to evaluate the confidence afforded by a public key. However their work has a static point of view and does not take into account the issue of public key revocation. In this paper, we show how to incorporate the revocation status information into the trust metrics for public key certification for the cases of both the single path and multiple paths certifying the same public key. To achieve our goal, we use a tailored form of a vector of trust model recently proposed. This would allow us to reason formally about when there is a need to check revocation status and how reliable the revocation mechanism should be in a given security application
Trusting Pseudonyms:
Anonymous Communication in Peer-to-Peer Reputation Systems
Michael Kinateder
Ralf Terdic
Kurt Rothermel
Full paper (PDF, 531Kb)
In this paper we present a novel approach to enable untraceable communication between pseudonyms. Our work provides strong sender and recipient anonymity by eliminating the need to know of each other’s address. We use a variation of Chaum mixes to achieve unlinkability between sender and recipient and introduce a concept called extended destination routing (EDR) which relies on routing headers constructed in multiple layers of encryption and published in a distributed hash table (DHT). In order to communicate, a sender requests from the DHT the recipient’s routing header, which is extended and used for routing the message via a mix cascade to this recipient. This work was performed in the context of the UniTEC reputation sys-tem and describes the functionality of its anonymous communication layer, which is completely independent of the other UniTEC layers. Al-though trust and reputation systems in general are typical application areas for our contribution, the presented concepts are suitable for various other application areas as well. We have implemented a prototype of UniTEC and present results from an ongoing evaluation in our network emulation testbed.
Object Trust Management for Information Quality Assurance in Virtual Organisations
Yanjun Zuo
Brajendra Panda
Full paper (PDF, 441Kb)
Ensuring information quality in a virtual organisation is crucial in order for the participants to confidently use imported data. Information quality is measured by the extent to which a user trusts the information as accurate. The component-based scheme presented in this paper assumes that several versions exist for a given object. Each version represents a subject’s opinion about the value(s) of the object, called object value(s). The presented model allows a user to evaluate the trustworthiness of a version of the given object based on how the object version was formed and how its components should be trusted. A label is associated with each object to provide available versions of the object and component information for each version. The trustworthiness of each component in our model can be used in calculating the primary trust value, which is calculated by direct experience of the evaluator, of a compound object version. The secondary trust value, calculated for each component using trusts assigned by other subjects to determine the final trust value of the compound object version, as explained in the paper, is more computationally efficient. But for that method to work, subject trust calculation is required. In addition to those methods for object trust calculation, heuristic approaches based on component information of a given object version are also presented to facilitate object trust evaluation.
Moleskiing.it: a Trust-Aware Recommender System for Ski Mountaineering
Paolo Avesani
Paolo Massa
Roberto Tiella
Full paper (PDF, 256Kb)
Recommender System (RS) suggests items to users they will like based on their past opinions. Collaborative Filtering (CF) is the most used technique and works by recommending to the active user items appreciated by similar users. However the sparseness of user profiles often prevents the computation of user similarity. Moreover CF doesn’t take into account the reliability of the other users. In this paper we present a real world application, namely moleskiing.it, in which both of these conditions are critic to deliver personalised recommendations. A blog oriented architecture collects user experiences on ski mountaineering and their opinions on other users. Exploitation of Trust Metrics allows to present only relevant and reliable information according to the user’s personal point of view of other authors trustworthiness. Differently from the notion of authority, we claim that trustworthiness is a user-centred notion that requires the computation of personalised metrics. We also present an open information exchange architecture that makes use of Semantic Web formats to guarantee interoperability between ski mountaineering communities
