Please use this identifier to cite or link to this item:
標題: 設計並檢視互動性個人化保險電子郵件推薦之有效性
Designing and Inspecting the Effectiveness of Interactive Personalized Insurance Recommendation by E-mail
作者: Chen, Yi-Chung
關鍵字: Insurance marketing;保險行;Personalized recommendation;Data mining;Flow Theory;E-mail Marketing;銷;個人化推薦;資料;探勘;沉浸理;論;電子郵件行;銷
出版社: 電子商務研究所
本研究依據所欲研究的問題,針對台灣地區20歲以上有使用電子郵件民眾為研究對象,採用雙重取樣法回收最終有效網路問卷130份,回收率為78.78 %。而研究方法包括次數分配(Frequency Distribution)、因素分析(Factor Analysis)、Cronbach’s α、結構方程式模型(Structural Equation Model)。本研究得出的結果如下:
(1) 透過資料探勘技術建立的保險個人化推薦系統能夠依據保險公司資料庫預測新顧客適合的保險推薦險種。
(2) 當顧客透過電子郵件瀏覽保險公司所提供的個人化保險推薦能夠產生沉浸效果。
(3) 顧客瀏覽個人化保險電子郵件推薦所產生的沉浸經驗對保險推薦效果有其直接的影響。

With the stability of information technology environment, the growth of Internet population and changes in customer needs have prompted life insurance companies to adopt their business models to grasp opportunities in the Internet area, and to increase more customers' purchase intention and loyalty. E- ail that has the characteristics including “Faster,” effective,” and “low cost,” has become one of marketing hannels. Recommendation system have been widely adopted in the nternet area, it can help to increase customers' purchase intention and loyalty. In this research, we adopt ecommendation technology in insurance area, and hope it can upport insurance company. Furthermore, “Flow Theory” is extensively applied in the Web environment and it can help to explain why people repetitiously use particular service. In this research, we adopt flow to inspect whether the personalized E-mail insurance recommendation work or not. According to the statement as above, in this research, we manipulate data mining technology to create a personalized E- ail insurance produces recommendation system, and we adopt the online questionnaire to find out the causal relationship among personalized E-mail insurance recommendation attributes, experiential flow, and personalized E-mail nsurance recommendation performances.
In this research, objects are people living in Taiwan, more han 20 years old and are used to use the E-mail. We adopt double sampling to collect 130 usable questionnaires and the esponse rate is 78.78 percent. The analysis methods in this research are analysis of frequency distribution, factor nalysis, Cronbach's α and structural equation model.
The important results of this research are as follows:
1. We can predict suitable insurance products to new customers with the data mining developed personalized insurance recommendation based on insurance company's database.
2. Customers are in the flow state when they reading the personalized E-mail insurance recommendation provided by nsurance company.
3. The flow experience has direct impact on the personalized E-mail insurance recommendation performances.
Appears in Collections:科技管理研究所

Show full item record
TAIR Related Article

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.