Intelligent Information Agent for B2C Electronic Commerce

• Main Authors: Yan-Hon Yen, 葉彥宏
• Other Authors: Chu-Chai Chan
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/87260719217348429456

碩士 === 朝陽科技大學 === 工業工程與管理系碩士班 === 90 === Abstract According to the research of National Information Infrastructure Enterprise Promotion Association, More than 8.3 million people use internet have exceeded at March, 2002, Internet and electronic commerce have became the most popular issues , also computers connect to the server of internet are over 570 thousand. As we have known internet have a big success in Taiwan. We can expect that the prospective environment of commerce will move toward electron market E-commerce. Internet not only changes human's life, the ways of working and thinking, but also goes to change theirs shopping behavior expenditure habits. The marketing of Internet is already growing up quickly; the whole world companies are desire to occupy their position in this newly market. Today Internet provides customer regular information but lack of personalized information, it waste customer a lot of time and cost. Only few websites actively user to provide their personalized information. Most of customer do not like to fill in information because it is in a hurry or can't find his/her own hobby. Therefore, the research develop an methodology to collect user's data and find out specific information for personalized homepage. From the point view of customer, website provides personalized information to save time in searching data . In this way, companion would provide user more professional service. The research will use intelligent agent to analysis user's preference. Of some unavailable data from customer, this research collect customer related to their behavior users' behavior and proceeds with Neural Networks Analysis to find out user's preference. The research classification four parts with attribution classification, behavior record, analyzed record and personalized homepage. 1.Attribution Classification: collect user's preference to do segmentation. 2.Behavior Record: record customer behavior. 3.Analyzed Record: use Neural Networks Analysis to find out customer favorite information or advertisement. 4.Personalized Homepage: Based on analysis ,personalized homepages are created.