A Spreading Activation Approach for e-Commerce Site Selection System

Abstract

This paper presents a dynamic associative network model for e-commerce site selection system based on the psycholinguistic theories of human memory; Spreading Activation Network (SAN). The system is designed to give personalized suggestions based on the user’s current personal preferences, other user’s common preferences, web-link structure, and activation rules. This work employs a SAN as a technique to provide the evaluation and selection mechanism that provides multiple parallel processes for perception by representing dynamic associations among web-links, user activities, and the relevance subjects of the websites. The system attempts to evaluate a number of sites in an unpredictable complex dynamic environment. Spreading activation explains the predictive top-down effect of knowledge. These processes select the group of the most suitable websites (context) in response to the current conditions (e-commerce activities) while the system continues working towards the user objective goal.