Fast and Secure Recommendations Protocols among Members of a Community of Agents

Abstract In recent years, the number of Internet users has increased very significantly. The utilization of the Internet for business purposes is everyday more factual. However this exponential growth of Internet users and networks has provoked an equivalent augment of malicious users. Hackers can take advantage of the low security level of the network, and hence, access to the valuable information exchanged through it. As a huge amount of electronic shopping choices become available, the reputation and interest of merchants and users have an economic value and probable costly consequences. Therefore, security mechanisms are needed for protecting the exchange of this type of information. When agents are representing users in Electronic Commerce scenarios, they are intended to cooperate with their trusted agents sharing recommendations about third parties, which represents the implementation of an electronic ‘word of mouth’. In this way, only the members of the group of agents should be authorized to access to its recommendations. This particular problem requires sharp responses according to computer performance, since it is supposed to be executed in real time. So typical approach –communications protection based on public key encryption- will be avoided in order to improve the efficiency. In this paper, we propose some security mechanisms to be applied in such specific domain: electronic communities of agents sharing recommendations about merchants. As each agent establishes its electronic community and this group of agents is short-lived and all of them will receive exactly the same messages, then we propose a protection based on the combination of hash chains and symmetric encryption.

[1]  Bruce Schneier,et al.  Applied cryptography : protocols, algorithms, and source codein C , 1996 .

[2]  John Riedl,et al.  Recommender systems in e-commerce , 1999, EC '99.

[3]  Stephen Hailes,et al.  Supporting trust in virtual communities , 2000, Proceedings of the 33rd Annual Hawaii International Conference on System Sciences.

[4]  José M. Molina López,et al.  A fuzzy model of reputation in multi-agent systems , 2001, AGENTS '01.

[5]  Tom Tedrick,et al.  Fair Exchange of Secrets , 1984, CRYPTO.

[6]  Stephen Marsh,et al.  Trust in Distributed Artificial Intelligence , 1992, MAAMAW.

[7]  Leonard N. Foner Clustering and Information Sharing in an Ecology of Cooperating Agents , 1995 .

[8]  David Chaum,et al.  Blind Signatures for Untraceable Payments , 1982, CRYPTO.

[9]  Tad Hogg,et al.  Enhancing privacy and trust in electronic communities , 1999, EC '99.

[10]  Manuel Blum,et al.  How to exchange (secret) keys , 1983, TOCS.

[11]  Pattie Maes,et al.  Kasbah: An Agent Marketplace for Buying and Selling Goods , 1996, PAAM.

[12]  Bikramjit Banerjee,et al.  Using bayesian networks to model agent relationships , 2000, Appl. Artif. Intell..

[13]  Leonard N. Foner,et al.  Yenta: a multi-agent, referral-based matchmaking system , 1997, AGENTS '97.

[14]  Adi Shamir,et al.  PayWord and MicroMint: Two Simple Micropayment Schemes , 1996, Security Protocols Workshop.

[15]  Neil Haller,et al.  The S/KEY One-Time Password System , 1995, RFC.

[16]  Giorgos Zacharia,et al.  Trust management through reputation mechanisms , 2000, Appl. Artif. Intell..