Engineering of Experience Based Trust for E-Commerce

Engineering of Experience Based Trust for E-Commerce

Zhaohao Sun (University Of Ballarat, Australia ), Jun Han (Beihang University, China), Dong Dong (Hebei Normal University, China ) and Shuliang Zhao (Hebei Normal University, China)
DOI: 10.4018/978-1-60566-669-3.ch014
OnDemand PDF Download:
No Current Special Offers


Trust is significant for sustainable development of e-commerce and has received increasing attention in e-commerce, multiagent systems (MAS), and artificial intelligence (AI). However, little attention has been given to the theoretical foundation and intelligent techniques for trust in e-commerce from a viewpoint of intelligent systems and engineering. This chapter will fill this gap by examining engineering of experience-based trust in e-commerce from the viewpoint of intelligent systems. It looks at knowledgebased trust, inference-based trust and their interrelationships with experience-based trust. It also examines scalable trust in e-commerce. It proposes a knowledge based model of trust in e-commerce and a system architecture for METSE: a multiagent system for experience-based trust in e-commerce. The proposed approach in this chapter will facilitate research and development of trust, multiagent systems, e-commerce and e-services.
Chapter Preview


Generally, trust is a positive belief or expectation about the perceived reliability of, dependability of and confidence in a person, an intelligent agent, organization, company, object, process, or system (Schneiderman, 2000). Castelfranchi and Tan (2001) assert that e-commerce can be successful only if the general public trust is established in the virtual environment, because lack of trust in security is one of the main reasons for e-consumers and e-vendors not to engage in e-commerce. Therefore, trust has received an increasing attention in e-commerce and information technology (IT). For example, Finnie and Sun (2007) investigate trust in e-supply chains. Olsson (2002) examines trust in e-commerce. Pavlou (2003) integrates trust with the technology acceptance model to explore the customer acceptance of e-commerce. Salam et al. (2005) examine trust in e-commerce and notice that “many customers may still not trust vendors when shopping online”. Wingreen and Baglione (2005) study the customer’s trust in vendors from a business viewpoint. Xiu and Liu (2005) propose a formal definition of trust and discuss the properties of trust relation. Xiong and Liu (2002) propose a formal reputation-based trust model by combining amount of satisfaction, number of interaction and balance factor of trust in a peer-to-peer e-community. However, the majority of studies are on trust in online purchase settings, whereas there is relatively less research on trust in e-commerce from a viewpoint of logic and intelligent systems.

Multiagent systems (MAS) have been successfully applied in many fields such as e-commerce (Sun & Finnie, 2004) and e-supply chain management (SCM) (Finnie, Sun & Barker, 2004; Finnie & Sun, 2007). MAS has also been used as a development methodology in many studies (Henderson-Sellers & Giorgini, 2005). Further, trust has drawn some attention in MAS. For example, Chen et al. (2005) propose a fuzzy trust model for MAS taking into account direct trust, recommendation trust and self-recommendation trust. Xiu and Liu (2005) discuss trust in distributed systems. Tweedale and Cutler (2006) discuss trust in MAS by proposing a trust negotiation and communication model for MAS architecture. Schmidt et al. (2005) apply a fuzzy trust model to an e-commerce platform. However, they have not examined engineering of trust in multiagent e-commerce system (MECS), which is of practical significance for multiagent e-commerce and e-services. This chapter will be devoted to engineering of trust and experience-based trust in MECS.

Experience-based reasoning (EBR) is a reasoning paradigm using prior experiences to solve problems, and could be considered an advanced form of knowledge-based reasoning (Sun & Finnie, 2007). This chapter will apply EBR to trust among intelligent agents within the MECS. In particular, the use of experience in establishing trust in other agents will be explored. Any organization has some history of dealing with problems relating to orders and perturbations in the network and the solutions applied, as well as some formal processes for dealing with these. To respond automatically, software must be capable of reacting as one would expect a human agent to do. The information available to the agent may come from a variety of sources, including analysis of historical information/experience at the information/planning level (Finnie & Sun 2007).

Key Terms in this Chapter

Knowledge-Based System: an AI application system mainly consisting of a knowledge base in a domain and inference engine. It is expected that everyday information systems will increasingly become knowledge-based and provide users with more assistance than they do today.

Multiagent Systems (MAS): have been studied in the field of distributed AI for more than 30 years (Weiss, 1999; Sun & Finnie, 2004). Recently, the term MAS is used for all types of systems composed of multiple agents showing the following characteristics (Sun & Finnie, 2004): Each agent has incomplete capabilities to solve a problem; There is no global system control over agents; Data are decentralized; Computation is asynchronous.

Expert system (ES): a special form of a knowledge-based system. Expert systems employ human knowledge to simulate expert performance, and they present a human-like facade to the users. Expert systems are one of the successful application fields in AI.

Fuzzy Logic: conceived by Lotfi Zadeh in 1965, is a mathematical technique for dealing with imprecise and fuzzy data/knowledge and problems that have many solutions rather than one. Although it is implemented in computers which ultimately make only yes-no decisions, fuzzy logic works with ranges of values, solving problems in a way that more resembles human logic. Fuzzy logic is used for solving problems with expert systems and other intelligent systems that must react to an imperfect environment of highly variable, volatile or unpredictable conditions (Zimmermann, 1996).

E-Commerce: already a discipline for research and development in commerce and information technology. The simple definition for e-commerce is doing business online. E-commerce can be also defined as the exchange of information, goods or services within business through the use of Internet technology (Sun & Finnie, 2004, p. 47). Usually, e-commerce and e-business is used interchangeably.

Case-Based Reasoning (CBR): a reasoning paradigm based on previous experiences or cases. As a system, a case-based reasoner solves new problems by adapting solutions that were used to solve old problems (Sun & Finnie, 2004). CBR is also a special form for experience-based reasoning (Sun & Finnie, 2007).

Experience-Based Reasoning (EBR): a reasoning paradigm based on human experience. The human EBR consists of eight inference rules at fundamental level, as mentioned in this chapter. Therefore, it is a logical basis for natural reasoning in particular, for natural intelligence in general.

Scalable Trust: motivated from the scalable system. Scalable trust of MAS can be defined as the trust that can fit for the system parameter changes, which cannot lead to the input’s remarkable increase and the output’s remarkable decrease (Zhao & Sun, 2008). Scalable trust consists of individual-level trust, network-based trust, group-based trust, organization-based trust and institution-based trust. Trust propagation from individual-level trust to group-based trust is important for MECS at the moment.

Trust: firm reliance on the integrity, ability, or character of a person or thing. In e-commerce, trust is the expectation that arises within a community based on commonly sharing norms from one member to another of that community. More generally, trust indicates a positive belief or expectation about the perceived reliability of, dependability of and confidence in a person, an intelligent agent, organization, company, object, process, or system (Schneiderman, 2000).

E-Services: a term for services on the Internet. E-services include e-commerce transaction services for handling online orders, application hosting by application service providers (ASPs) and any processing capability that is obtainable on the Web. There is a trend to integrate e-commerce and e-business with e-services.

Complete Chapter List

Search this Book: