Attribute selection with application to recommender systems for green logistics e-Commerce

作者：Yang Jaekyung; Choi Myoungjin; Yu Wooyeon^*

来源：Asia Life Sciences, 2015, 511-524.

• e-Commerce
• attribute selection
• data mining
• green logistics
• recommender

Along with emergence of green (or reverse, healthcare) logistics considering the environment, a variety of technologies has been developed. The recommender system for an on-line auction system for the green logistics is one of these systems. In this paper, we propose a new nested partition-based attribute selection method that can be used to make recommendation rules simpler to auction users. This attribute selection method involves a process for determining which attributes are relevant in that they predict or explain the data. Further, it may improve scalability and make interpreting recommendation rules (learning models) easier. In order to overcome the limitation of the nested partition attribute selection method where it can originally be adapted to discrete type of attributes, two attribute quality evaluators are introduced with experiment results in order to deal with the mixed type of data. As a case study, a recommender system that can be effectively used in a business to business,e-Commerce for green logistics is provided using classification rules and the new attribute selection method. The systems create recommendation rules for auction bidding.

• 出版日期2015-7