Enhancing Web-Based Configuration with Recommendations and Cluster-Based Help

In a collaborative project with Tacton AB, we have investigated new ways of assisting the user in the process of on-line product configuration. A web-based prototype, RIND, was built for ephemeral users in the domain of PC configuration. Two mechanisms were added to a commercial configurator produced by Tacton: i) automated recommendations that display social trails associated with the configuration; and ii) a help system based on term clustering. Recommendations based on previous customer selections are made on separate attributes as well as full configurations, i.e. complete PCs. The early rater problem is solved using a probabilistic bootstrapping approach. The help system supports novice users browsing for help information, as well as experienced users able to pose exact queries.