中文摘要：

采用一套基于邻域设施的模糊区位取向方法和模糊多属性决策方法,针对中国北京、上海、深圳、天津4城市淘宝网站6大商品类别网络店铺的空间集聚情况,通过其所属的完全、重叠、过渡和不完全4种区位类型,揭示区位取向的倾向性特征和信息时代区位影响因素的变化,并进一步探讨区位取向明显的数码类别商品的网络店铺的最优区位地选择。形成如下认识：①不同商品类别情况下区位取向有很大不同,分散与集聚共同存在。集聚只发生在贵重商品如数码类别商品的网络店铺上,其仍依赖于传统邻域设施;而绝大多数普通商品则相反,几乎不存在重叠区位的布局,已经表现出一种不再遵循传统区位地理论的新的区位取向规律。②不同商业模式下区位取向不同,c2c电子商务对其影响更趋向于不同于传统的区位选择,表明c2c电子商务是网络店铺区位取向影响因素变化的基础。除了旧有的传输网络、货源供给、市场集聚因素外,技术劳动力因素、信任因素渐成主流。③利用三角模糊数能计算出基于邻域设施集聚的、区位取向明显的数码类别商品网络店铺的最优区位。选择最优区位是从重叠区位向完全区位过渡的。④在电子时代传统的区位理论将得到一定补充或纠正才能全面正确地解释电子服务网点的区位现象。

英文摘要：

This paper employs two basic approaches： fuzzy location-orientation method based on neighborhood facilities and fuzzy multiple attribute decision making （MADM） method. The study focuses on spatial agglomeration analysis of e-shops which have tangible stores at Taobao website. The Taobao e-shops located in Beijing, Shanghai, Shenzhen and Tianjin have six categories of goods： digital products, household collections, clothing, skin care, stationery and sports. These e-shops can be divided into four location types： complete, overlapping, transitional and partial. Based on these four types, this paper reveals the biased characteristics of location-orientation of various e-shops and the changes of location orientation influence factors in information era, and further explores the optimal location choices of digital product e-shops which have obvious location distribution. The research findings are as follows. （1） Different goods categories have significant differences on e-services network location choice （ESNLC）, and decentralization and agglomeration coexist. Agglomeration only happens at precious commodities such as digital categories and it still relies on traditional neighborhood facilities. By contrast, the majority of ordinary commodity e-shops barely have overlapping location layout, and have already shown a new location-oriented regular pattern, which no longer follows the traditional location theory. （2） The e-services network location choices under different commercial models differ from each other. The influences on ESNLC of c2c e-commerce model tend to be more different from the traditional location choice, which indicates that c2c e-commerce is the foundation of changes of location-orientation influence factors of e-shops. In addition to the old transmission network, sources of supply, market concentration factors, skilled labor factors and trust factors would become the mainstream. （3） Triangular fuzzy numbers can be used to calculate the optimal location of precious goods e-sho