Catálogo de publicaciones - libros

We analyze a mechanism that provides strong incentives for the submission of truthful feedback in virtual communities where services are exchanged on a peer-to-peer basis. Lying peers are punished with a severity that is exponential to their frequency of lying. We had first introduced and evaluated experimentally the mechanism in [1]. In this paper, we develop a Markov-chain model of the mechanism. Based on this, we prove that, when the mechanism is employed, the system evolves to a beneficial steady-state operation even in the case of a dynamically renewed population. Furthermore, we develop a procedure for the efficient selection of the parameters of the mechanism for any peer-to-peer system; this procedure is based on ergodic arguments. Simulation experiments reveal that the procedure is indeed accurate, as well as effective regarding the incentives provided to participants for submitting truthful feedback.

Autonomous entities in artificial societies are only willing to cooperate with entities they trust. Reputation systems keep track of the entities’ behavior and, thus, are a widely used means to support trust formation. In a P2P network, the reputation system needs to be distributed to the individual entities. In previous work, we have shown that some of the limitations of distributed reputation systems can be overcome by making use of hard evidence. In this paper, we take this idea one step further by deriving beliefs of others’ trustworthiness from one’s own experiences and the available hard evidence. For this purpose, we justify why a self-interested autonomous entity may choose to behave according to the norms of the system designer. As a consequence, the proposed belief model does not only incorporate behavioral beliefs but also beliefs regarding the normativeness of an entity. We prescribe how beliefs are revised if new evidence becomes available. The introduced models for recommendations and belief formation enable us to prove that self-interested entities always issue truthful recommendations regarding transactional behavior. The simulative evaluation shows that a self-interested entity can be expected to be normative and, thus, to comply with our system design.

Since there is no method to verify the trustworthiness of shared files in P2P file sharing systems, malicious peers can spread untrustworthy files to the system. In order to prevent untrustworthy files from spreading, we propose an effective trust management scheme using peer reputation and file reputation together in a DHT-based structured P2P systems. Simulation results show that the proposed scheme effectively restrains the spreading of untrustworthy files even in cases where malicious peers change their identities. Simulation results show that the overall message cost for managing trust data is relatively low. We also propose a replication scheme so as to avoid the loss or corruption of trust data.

can be powerful tools for promoting exchanges and building sustainable relationships among selfish peers on the Internet.

-WAT[1] is a proposed such currency based on the WAT System, a polycentric complementary currency using as its media of exchange. Participants spontaneously issue and circulate the tickets as needed, whose values are backed up by chains of trust. -WAT implements the tickets electronically by exchanging messages signed in OpenPGP.

This paper claims that the design of -WAT is incentive-compatible as to protection against moral hazards, or threats caused by selfish peers because they may take advantage of the rules; such hazards are defused in -WAT if the participants react against misbehaviors of others by pursuing their own benefits.

A reference implementation of -WAT has been developed in the form of an XMPP instant messaging client. We have been putting the currency system into practical use since June 2004.

In reputation-based peer-to-peer systems, reputation is used to build trust between peers and help selecting the right peers to download from. In this paper, we argue that reputation should not be used for service differentiation among the peers. To provide the right incentives for peers to share files and contribute to the system, the new concept of is introduced for partially decentralized peer-to-peer systems. Service differentiation is achieved based on the of the peers rather than their reputations. Simulation results assess the ability of the proposed algorithm to effectively identify free riders and malicious peers that upload malicious content, hence reducing the level of service provided to these peers and preserving network resources. On the other hand, good peers that contribute to the system receive better services which increases their satisfaction significantly.

The research in the paper is motivated by building a decentralized/P2P XML storage on top of a DHT (Distributed Hash Table). The storage must provide high data availability and support updates. High data availability in a DHT can be guaranteed by data replication. However, DHTs can not provide a centralized coordination guaranteeing data consistency upon updates. In particular, replicas may have different values due to concurrent updates or partitioning of the P2P network. An approach based on versioning of replica values is presented proposing a decentralized concurrency control system, where probabilistic guarantees can be provided for retrieving a correct replica value. This paper presents the protocol as well as a statistical analysis of the lower bound of the probabilistic guarantees.

Structured overlay networks for Peer-to-Peer systems (e.g. based on Distributed Hash Tables) use proactive mechanisms to provide efficient indexing functionality for advertised resources. The majority of their occurrences in proposed systems (e.g. Chord, Pastry) provide upper bounds (logarithmic complexity with respect to the size of the graph representing the network) on the communication cost in worst case scenarios and their performance is superior compared to unstructured alternatives. However, in particular (empirically observed) scenarios where the popularity of the advertised resources follows a distribution considerably different from the uniform distribution, structured P2P networks may perform inferiorly compared to well designed unstructured P2P networks that exploit effectively the resource popularity distribution. In order to address this issue, a very simple caching mechanism is suggested in this paper that preserves the theoretical superiority of structured overlay networks regardless of the popularity of the advertised resources. Moreover, the churn effect observed in Peer-to-Peer systems is considered. The proposed mechanism is evaluated using simulation experimesnts.

Achieving interoperability and automation in job execution is of utmost importance for next generation e-Commerce applications. This paper proposes a design that integrates three prominent technologies: intelligent software agents, peer-to-peer networking and the Semantic Web. SeMPHoNIA is an architecture for an agent-based marketplace, utilizing knowledge from RDF product repositories, in an open peer-to-peer environment. The platform defines the basic stages of the process of e-trading, facilitating users in closing deals in automated manner. The implementation of our approach is demonstrated in the context of auction scenarios. A performance evaluation of the system is also presented.

This paper describes a multi agent system (MAS) for distributed composition of Semantic Web services. Since our system is intended to function in highly dynamic environments, where heterogeneous agents rapidly join and leave the system, we consider P2P approach as most suitable for facilitating agent and service discovery. The MAS is based on Chord P2P network, which allows the MAS to dynamically publish and locate available Semantic Web services, which are specified with OWL-S. In order to compose new Web services from existing ones, agents apply symbolic reasoning in a cooperative problem solving manner.

This paper focuses on the latency reduction problem in massively multiplayer games (MMGs). As the client-server (CS) architecture in use today in most commercial MMGs applications are exposing its weakness in scalability as the number of players increases, researchers start to consider the peer-to-peer (P2P) model that has inherent high scalability for MMGs. However, existing P2P models generally lead to high latency that significantly detracts from the playing experience. To improve this, we present a novel communication model intended to reduce the latency of network communication on game states. The model, based on DHT protocol, can select the best candidate node to perform server’s role for a game zone. In contrast, existing P2P models designate ’peer servers’ randomly without considering latency effects. The simulation shows that our approach obtains much lower latency than existing P2P models. The performance of our approach is shown to be even better than that of CS models.