A Framework for Dynamic Facet Ordering Search in E-Commerce

Abstract

Faceted browsing is wide employed in internet outlets and merchandise comparison sites. In these cases, a hard and fast ordered list of aspects is usually utilized. This approach suffers from 2 main problems. First, one has to invest a major quantity of time to plan an efficient list. Second, with a hard and fast list of aspects it will happen that a aspect becomes useless if all product that match the question area unit associated to it specific aspect. during this work, tend to gift a framework for dynamic aspect ordering in e-commerce. supported measures for specificity and dispersion of aspect values, the totally automatic algorithmic program ranks those properties and aspects on prime that result in a fast drill-down for any attainable target product. In distinction to existing solutions, the framework addresses e-commerce specific aspects, like the likelihood of multiple clicks, the grouping of aspects by their corresponding properties, and therefore the abundance of numeric aspects. in an exceedingly large-scale simulation and user study, approach was, in general, favourably compared to an aspect list created by domain specialists, a greedy approach as baseline, and a progressive entropy-based resolution.