PROJECT TITLE :

Dynamic Facet Ordering for Faceted Product Search Engines - 2017

ABSTRACT:

Faceted browsing is widely employed in Web retailers and merchandise comparison sites. In these cases, a fastened ordered list of sides is typically employed. This approach suffers from two main problems. 1st, one wants to take a position a vital amount of time to plan an efficient list. Second, with a fastened list of facets, it can happen that a aspect becomes useless if all product that match the question are associated to that specific side. In this work, we present a framework for dynamic aspect ordering in e-commerce. Primarily based on measures for specificity and dispersion of side values, the absolutely automated algorithm ranks those properties and facets on high that result in a quick drill-down for any possible target product. In contrast to existing solutions, the framework addresses e-commerce specific aspects, such as the possibility of multiple clicks, the grouping of aspects by their corresponding properties, and also the abundance of numeric facets. In a very massive-scale simulation and user study, our approach was, normally, favorably compared to a aspect list created by domain consultants, a greedy approach as baseline, and a state-of-the-art entropy-primarily based resolution.