A mental model approach for category hierarchy maintenance on sellers' self-input items in e-commerce websites

This paper proposes a mental model approach to support category hierarchy maintenance on sellers' self-input items in e-commerce websites. We conduct co-occurrence analysis between sellers' self-input items and existing website category items, and use Hierarchical Clustering Analysis to explore hierarchical structure and the association relationships hidden among these subjects. At last, we draw on Multidimensional Scaling to visualize and validate the results of Hierarchical Clustering Analysis. So the corresponding positions of different sellers' self-input items in the existing category hierarchy can be determined. An empirical study has also been undertaken with the log data of one Chinese e-commerce website. The achievements of this paper can be applied to determine the most proper category hierarchy or recommend the top N categories for sellers' consideration when they list items on e-commerce websites.