A hybrid CBR-ANN approach to the appraisal of internet domain names

Abstract

Good domain names have become rare and trading with premium domain names has developed into a profitable business. Domain appraisals are required for many different reasons, e.g., in connection with a loan on a domain name. The aim of this paper is to analyze various methods for estimating prices for domain names. The criteria for this are predictive accuracy, traceability and speed of the appraisal. First, the scientific relevance of the topic is demonstrated based on intensive literature and Internet research. Several approaches based on artificial neural networks (ANNs) and case-based reasoning (CBR) are developed for estimating domain name prices. In addition, hybrid appraisal approaches are introduced that are built up on CBR and which use ANN for improved adaptation and similarity determination. The approaches are evaluated in several configurations using a training set of 4,231 actual domain transactions, which demonstrates their high usefulness.