Reserve Price Recommendation by Similarity-Based Time Series Analysis for Internet Auction Systems

It is very important that sellers provide reasonable reserve prices for auction items in internet auction systems. We recently have proposed the agent based system in order to generate reserve prices automatically, based on the case similarity of information retrieval theory and the moving average of time series analysis. However, the approaches have some problems that they can not reflect the recent trend of auction prices on the generated reserve prices properly, because they only provide the bid prices or average prices of items, by using the past auction data for sellers. In this paper, in order to overcome the problems, we propose Similarity-Based Time Series Analysis which search the past bid price through case similarity and then generate reserve prices by using time series analysis. Through performance experiments, we show that the successful bid rate of the new method can be improved by preventing sellers from making unreasonable reserve prices compared with the past methods.