Familiarity and Trust: Measuring Familiarity with a Web Site

This work aims at measuring familiarity to contribute to the formalization of trust. Trust has always been bundled with familiarity to become a popular topic in the areas of psychology, sociology and computer science. Correlation between familiarity and trust has been explored and proved by many studies from different perspectives. A new model of trust has been proposed by Carter and Ghorbani to formalize the value-centric trust in agent societies. However, the measurement of familiarity in their work is roughly the similarity of values between two agents. Familiarity measurements proposed by other researchers are not convenient due to the instability and abstruseness of familiarity, or are useful only in certain circumstances and are quite problem-specific. We propose a convenient way of measuring familiarity with a Web site and continuously updating its value based on the exploration of factors that may affect familiarity. The five major factors include prior experience, repeated exposure, study duration, level of processing and forgetting rate. The human factors are mapped to the properties of Web application domain through a factors hierarchy. Experiments to evaluate the performance of the proposed measurement are discussed in the future work section.