An application of web-based data mining: selling strategies for online auctions

Purpose - This study aims to introduce an application of web-based data mining that integrates online data collection and data mining in selling strategies for online auctions. This study seeks to illustrate the process of spider online data collection from eBay and the application of the classification and regression tree (CART) in constructing effective selling strategies. Design/methodology/approach - After developing a prototype of web-based data mining, the four steps of spider online data collection and CART data mining are shown. A business dataset from eBay is collected, and the application to derive effective selling strategies for online auctions is used. Findings - In the web-based data-mining application the spiders can effectively and efficiently collect online auction data from the internet, and the CART model provides sellers with effective selling strategies. By using expected auction prices with the classification and regression trees, sellers can integrate their two primary goals, i.e. auction success and anticipated prices, in their selling strategies for online auctions. Practical implications - This study provides sellers with a useful tool to construct effective selling strategies by taking advantage of web-based data mining. These effective selling strategies will help improve their online auction performance. Originality/value - This study contributes to the literature by providing an innovative tool for collecting online data and for constructing effective selling strategies, which are important for the growth of electronic marketplaces.