| Summary: | 碩士 === 朝陽科技大學 === 資訊管理系碩士班 === 92 === A powerful online recommendation system in Electronic Commerce (EC) must know its target customers well and employ effective marketing strategies. Market research is a very important way to know the customers well. For high-tech products with great variety such as computers, cellular phones, and digital cameras, customers’ knowledge level towards products may have a decisive influence on their purchase decision. While many on line recommendation systems focus on utilizing data mining techniques in user profile and transaction data, this paper presents a method for recognizing customer knowledge level as a preprocess for more effective on line recommendation in EC. The method consists of two Back Propagation Networks and predicts based on customer characteristics and online navigation behaviors. A simple simulated digital camera EC store case study was conducted and the good preliminary results implies the good potential of the proposed method.
|
|---|
