Design and evaluation of improvement method on the Web information navigation - a stochastic search approach

Abstract

With the advent of fast growing Internet and World Wide Web (WWW), more and more companies start the electronic commerce to enhance the business competitiveness. On the other hand, more and more people surf on the Web for information gathering/processing. Due to unbalanced traffic and poorly organized information, users suffer the slow communication and disordered information organization. The information provider can analyze the traffic and uniform resource locator (URL) counters to adjust the organization; however, heterogeneous navigation patterns and dynamic fluctuating Web traffic make the tuning process very complicated. Alternatively the user may be provided with guidance to navigate through the Web pages efficiently. In this paper, a Web site was modeled as a Markov chain associated with the corresponding dynamic traffic and designated information pages. We consider four models: inexperienced surfers on guidance-less sites, experienced surfers on guidance-less sites, sites with the mean-length guidance, and sites with the known-first-arc guidance (generalized as sites with dynamic stochastic shortest path guidance). Simulation is conducted to evaluate the performance of the different types of navigation guidance. We also propose a reformulation policy to highlight the hyperlinks as steering guidance. The evolution on complexity and applicability is also discussed for the design guideline of general improvement methods. The paper concludes with the summary and future directions.