Trust and reputation for service-oriented environments: Technologies for building business intelligence and consumer confidence

Abstract

Trust has played a central role in human relationships and hence has been the subject of study in many fields including business, law, social science, philosophy and psychology. It has played a pivotal role in forming contracts, carrying out business, assisting people work together co-operatively and underpins many forms of collaboration. Closely related to this notion of trust is the concept of reputation within a community, with other peers or society in general. This is frequently used as the basis of a judgement as to whether to trust an individual or organisation particularly in the absence of previous direct contact with them.These two concepts of trust and reputation have assumed increasing importance in the information technology field.We note that increasingly in the networked economy, communication, collaboration, research, education, transactions, sales, marketing and industrial system controls are being conducted over the internet. As a result, trust becomes crucial to the enterprise or corporation, business, service provider, government and consumer making a decision to carryout an interaction on the internet. Lack of trust between communicating parties, results in a situation that is popularly known as 'prisoner's dilemma', in which either of the two parties could resort to unfair practices. The two parties may be involved in transactions worth thousands of dollars, but it may not be possible to detect any unfair practices until the outcome of the transaction. Time is needed for the two communicating parties to develop trust in each other before the business transaction occurs. It has become a significant element of business intelligence.This book aims to increase the level of understanding of the Trust and Reputation body of knowledge for service-oriented environments.The scope of this book includes provision of:a)A detailed explanation of the concepts of trust, trust relationships, trustworthiness, trust value, trustworthiness value, trustworthiness measure, reputation, reputation value and reputation measure and trustworthiness prediction;b)A detailed analysis of the dynamic natures of trust and reputation as well as the context specific and time dependent natures of trust and reputation;c)A clear understanding of Trust and Reputation Ontologies;d)A detailed methodology for establishing trust and assigning trustworthiness through direct interaction or through 3rd party recommendation and reputation, and trustworthiness prediction through historical datasets and reputation values;e)A detailed trust modelling technique;f)Advice on business intelligence through the use of trust and reputation technology and providing a clear distinction between existing business intelligence tools and this new class of technology in addition to identifying how they offer business intelligence and why;g)A clear distinction between web services and service-oriented environments, from a technology perspective and from business perspectives; andh)A clear distinction between trust and security, and how to build business intelligence and consumer confidence through trust.The service-oriented network environment can be viewed in three types of networked environments. The first type are centralized networks such as Client-Server systems. The second type are decentralized networks such as peer-to-peer (P2P) or Grid (a big-networked virtual computer) or Mobile Networks. The third type are partially centralized and partially decentralized networks that are a combination of both client-server and server-to-server (peers) infrastructures such as an Ad-Hoc network. The challenge, in the above three types of service-oriented environments, is due to the fact that the communicating parties or agents can carry out interactions in one of three ways:*Anonymous (identity is unknown during communication)*Psuedo-anonymous (known by a false identity during communication)*Non-anonymous (identity is known during communication)However, unlike the centralized client-server, which enables non-anonymous communication (where trust is pre-established before communication) distributed communication has each server with its own trust management system.It is difficult to establish trust between two communicating agents in a distributed network as most of the time these operate in an anonymous environment. Without trust relationships, an agent will be reluctant to carry out an interaction with another agent.Service-oriented networked environments have assumed new significance with the advent of web services.There is a common misconception that the studies of Trust and Security are the same thing.Security in computing field refers to the process of enabling secure communication between two communicating parties or machines in the service-oriented network environment, whereas trust is the belief that the Trusting Agent has in another agent?s willingness and capability to deliver Quality of Service, and it is context and time dependent. Trust is fundamental to business, commerce, education, politics and any form of collaboration. The more rationalized trust we have, the lower the risk of interaction we achieve. Conversely, the lower the rationalized trust we have, the higher the risk we encounter. This book aims to integrate concepts from economics, computing, science, engineering, marketing, business, law, sociology and psychology to address the need for trust in service-oriented networked environments.The emerging Trustworthiness Technologies are re-sharping the networked economy by providing business intelligence and consumer confidence through the creation of open, fair, convenient and transparent service-oriented environments for business transactions, Quality of Services, reputation of sellers and products as well as loyalty of buyers and their accountability. The Trustworthiness Technology acts as:*A quality assessor for businesses to fulfil their contractual obligations including Quality of Service and product delivery;*A consumer watch-dog for deterring unfair trading, dishonest dealing, defects in products and discriminatory processes;*An interaction mediator that helps distributed network buyers, sellers and online users to communicate, build trust relationships and create an opportunity for business collaboration and competition;*A strategic advisor for both online and off-line business providers to capture customer needs, competitors' operations, marketing trends etc;*A brand advertiser for the trustworthy service providers to assist with maintaining their Quality of Service and retaining their customers; and*A safeguard for all customers and all agents in the service-oriented networked environment to remove Untrusted Agents (defining within this text), malicious service providers and to guard against their attacks.This is a new-age, important, frontier technology for business intelligence. Study of this technology is applicable to computer science, computer engineering, software engineering and network engineering. It is also well suited for business, IT (Information Technology), IS (Information Systems) and eCommerce (Electronic Commerce) courses. It is a booming and important area for research, teaching and commerce. Security is of paramount importance for most countries. However, Trust plays a key role in security (in the broader sense) as it reduces the risk. We also see that the latest government funding has provided a huge amount of support for security and trusted platforms, programs and projects and many universities have newly created degrees in Security. Despite this interest, there is not a lot of material available on Trust and Reputation particularly in the newly emerging service-oriented environments. In the last couple of years many worldwide prestigious conferences and workshops (IEEE and ACM) incorporated 'Trust' into their tracks or themes. Several groups run specific workshops or conferences specifically on Trust. New subjects and fresh courses are appearing on topics such as Trust and Reputation. It is envisaged that trust management systems will become a commodity on any computer, server, network or business over the next 5 to 10 years (Dellarocas, 2004; Alexa, 2004).