Optimal design of link structure for e-supermarket website

The objective of this research is to optimize the link structure of webpages for an e-supermarket. Customers are looking for greater convenience in shopping when they visit the website of an e-supermarket, while e-supermarket managers prefer webpages that contain information about profitable products to be visited more frequently. In order to balance the interests of both parties and to aid the webmaster in updating the website regularly, we present a mathematical model with the objective of minimizing the overall weighted distances between webpages. An updating algorithm is used to determine the distance between pages. It is proved to be more efficient under certain special circumstances. We propose the statistical Hopfield neural-network and the strategic oscillation-based tabu-search algorithms as solving methods. The former is appropriate for optimizing small-scale problems. The latter is good at solving large-scale problems approximately. The preliminary validity of the model and the performance of the algorithms is demonstrated by experiments on a small website and several large websites, using randomly generated data. The destination pages that customers and website managers preferred are proved to be more accessible after optimization.