Modern business systems are Internet-based and require innovative ways to carry out business transactions autonomously and efficiently. This special issue of Electronic Commerce Research is devoted to agent-based models, techniques, and applications for building Internet e-business systems. It consists of eight papers derived from the works which were introduced in selected papers of the international workshop "Agents for E-Business on the Internet" in Las-Vegas, 2001.

The papers in this issue deal with different aspects of e-business with a common goal: to show that agent technology is a promising tool to conduct business via Internet in an automatic, intelligent, and effective way. We will address some of the most important issues of e-business such as: e-trust/security, e-pricing, e-negotiation/e-argumentation, e-trading, and e-management.

Trust is the foundation of business partnership and is especially critical in such open business environment as the Internet. Simulating trust and handling E-trust is a very interesting and challenging problem. The paper by Chihiro Ono, Satoshi Nishiyama, Keesoo Kim, Boyd C. Paulson, Jr., Mark Cutkosky, Charles J. Petrie Jr. brings e-trust and E-partnership into agents by providing a human-like trust facilitator which takes advantage of informal information such as the words-of-mouth to build trusts between partner-agents. In a more practical approach, Yuh-Jong Hu�s paper focuses on building a system to handle agent trusts via digital certificates. The author has developed protocols for agent authentication, authorization, delegation and shows how they can be applied for C2B, B2B, and C2C business models.

In Internet-based business, E-pricing is another important issue since distance is no longer a concern and prices become almost the only fact that dominates the user decision in purchasing goods or services. In order to define "the right price at the right time" Joan Morris, Amy Greenwald, and Pattie Maes�s paper provides a tool named Learning Curve Simulator to analyze agent pricing strategies in the e-marketplaces so that the selling agents can use the suitable pricing strategies in various situations. In search for a flexible customer profiling and pricing mechanism, Prithviraj Dasgupta and Michael Melliar-Smith�s paper proposes a pricing system which customizes the prices based on customers preferences.

Just like in the traditional business, everything in e-business on the Internet should be negotiable. Nevertheless, how can we build software agents which are able to do what is not easy even for human businessmen? Hajime Sawamura, Masashi Yamashita, and Yuichi Umeda�s paper shows how dialectical reasoning can be applied and embedded in agents to make them able to persuade the Internet customers through reasonable argumentations, just as the human salesmen would do in the real word.

Trading is the most active form of business. E-trading challenges us by its large scale, dynamic, distributed, and diverse information. Darryl N. Davis, Yuan Luo, and Kecheng Liu�s paper describes an agent-based system which uses KADS and Zeus for stock trading and personal portfolio management on the Internet where the data are changed frequently and the decisions need to be made within very short periods of time.

E-management is a key issue in building efficient e-business systems. Eugene Santos Jr., Feng Zhang, and Peter B. Luh�s paper addresses the problem of logistics management for agent-based business systems where the goal is to achieve overall efficiency in intra-organization competitions. Unlike other agent-based approaches the authors propose an interesting idea of using Lagrangian relaxation to break down this complicated problem.

Among intermediary business activities, mapping business demands and supplies occupies a central place. To achieve business automation, Hanh Pham, Yming Ye, and Vien Nguyen�s paper generalizes the mapping problems for both e-management and e-trading with multiple criteria and multiple values. As the demands and supplies in Internet-based systems may be hidden in different forms and locations the authors develop invisible Internet agents called knowledgeable objects for carrying and mapping heterogeneous and distributed business data in a reactive and autonomous way via agent bidding.

With these papers we hope to bring more experiences and ideas which can be used for future research as well as for building practical applications. This special issue came about thanks to the extraordinary efforts from all the authors and thanks to the kind encouragement from professor Bezalel Gavish. I also would like to express our special gratefulness and appreciation to Morad Benyoucef, Prithviraj Dasgupta, Darryl Davis, Andrew Davenport, Joan Morris DiMicco, Chirakkal Easwaran, Yuh-Jong Hu, Benoit Leloup, Nadege Marchand, Eugene Santos, Hajime Sawamura, Michael Stroebel, Jeffrey Teich, and other reviewers for their precious time and their valuable comments.

Article ID: 5122979