网络购买决策关键影响因素挖掘研究

发布时间：2018-06-20 13:04

本文选题：网络购买 + 数据挖掘　；参考：《东华大学》2011年硕士论文

【摘要】：随着以电子信息技术为基础的电子商务在国内的快速发展,网络购物已经成为中国网民工作生活的一部分,网上购物成为一种新型的商业模式已经逐渐被中国消费者接受,并逐步成为一种日常化的购物方式。这也吸引了越来越多的专家学者把其作为研究对象,其中对于网络环境下消费者的消费心理、动机和行为等方面有较多论述研究,然而对于网络消费者购买决策影响因素的研究并不多见。本研究在总结和吸取以往消费者购买相关研究成果的基础上,结合当前电子商务与网络营销理论实践的新发展,通过实验与实证相结合的研究方法,研究我国网民网络购买决策的典型影响因素,通过数据挖掘方法对其进行研究。先通过系统聚类算法从47个网络购买决策的影响因素中聚类出网站个性化功能定制、网站娱乐性、网站网址记忆容易、有网络广告宣传、店铺交易信息公开透明、店铺商品为知名品牌、店铺信誉7个因素是网络购买的典型影响因素,然后利用决策树算法通过实验研究方法对此7个典型影响因素进行挖掘,挖掘出各个影响因素的重要性大小。以期研究成果可以为网络卖家已经相关研究领域提供参照与借鉴。本文的创新点主要体现在以下两个方面：内容创新：目前,国内外学者对于网络环境下消费者的消费心理、动机和行为等方面有较多论述,然而对于网络消费者购买决策关键影响因素的研究并不多见。把网络购买决策影响因素作为本研究的对象,可以说是对该领域研究的一个探索、一个尝试。目前少有的关于网络购买决策影响因素的研究中,大多是针对网络购买决策影响因素中的一个或数个进行研究,本研究尝试着能系统地对网络购买决策影响因素进行整体性研究,更具合理性和接近实际。方法创新：R型系统聚类算法常应用于解决复杂的、难以量化的且相关联的多个因素间的聚类问题。本研究先通过R型系统聚类算法对网络购买的所有影响因素进行聚类,筛选出网络购买决策典型影响因素,以此为基础,设计实验获取实验数据作为决策树数据挖掘算法的输入,得到对网络购买决策典型影响因素按重要性的排序。
[Abstract]:With the rapid development of electronic commerce based on electronic information technology in China, online shopping has become a part of the working life of Chinese Internet users, online shopping has become a new business model has been gradually accepted by Chinese consumers. And gradually become a daily way of shopping. This has also attracted more and more experts and scholars to take it as the object of study, among which there are more exposition and research on consumer consumption psychology, motivation and behavior in the network environment. However, the research on the influencing factors of online consumer purchase decision is rare. On the basis of summing up and absorbing the previous research results of consumer purchase, combined with the new development of electronic commerce and network marketing theory and practice, this study combines the experimental and empirical research methods. This paper studies the typical influencing factors of purchase decision of Internet users in China, and studies them by data mining method. Firstly, through the systematic clustering algorithm, we cluster out the personalized function customization of the website, the entertainment of the website, the easy memory of the website address, the publicity of the network advertisement, the transparency of the shop transaction information, among the influencing factors of 47 network purchase decisions. Store merchandise is a well-known brand, and store reputation is a typical influence factor of network purchase. Then the decision tree algorithm is used to excavate the seven typical influencing factors through experimental research method. Dig out the importance of each factor. With a view to the research results for online sellers have been related to provide reference and reference. The innovation of this paper is mainly reflected in the following two aspects: content innovation: at present, domestic and foreign scholars have more exposition on consumer consumption psychology, motivation and behavior in the network environment. However, there are few researches on the key factors of online consumer purchase decision. Taking the influencing factors of network purchase decision as the object of this study can be said to be an exploration and an attempt in this field. At present, most of the few researches on the influencing factors of network purchase decision are aimed at one or more of the influencing factors of network purchase decision. This study attempts to systematically study the influencing factors of network purchasing decision, which is more reasonable and close to reality. Methods the innovation of class R clustering algorithm is often applied to solve the complex, difficult to quantify and related multiple factors clustering problem. In this study, we first cluster all the influencing factors of network purchase through R-type system clustering algorithm, and screen out the typical influencing factors of network purchase decision. The experimental data are obtained as the input of the decision tree data mining algorithm, and the order of the typical influencing factors of the network purchase decision is obtained according to the importance of the data mining algorithm.
【学位授予单位】：东华大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：F724.6;F224

【引证文献】