1、中文 3610 字 毕业论文（设计） 外文翻译 外文原文 Impact of Social Influence in E-Commerce Decision Making ABSTRACT Purchasing decisions are often strongly influenced by people who the consumer knows and trusts. Moreover, many online shoppers tend to wait for the opinions of early adopters before making a purchase decis
2、ion to reduce the risk of buying a new product. Web-based social communities, actively fostered by E-commerce companies, allow consumers to share their personal experiences by writing reviews, rating others reviews, and chatting among trusting members. They drive the volume of traffic to retail site
3、s and become a starting point for Web shoppers. E-commerce companies have recently started to capture data on the social interaction between consumers in their websites, with the potential objective of understanding and leveraging social influence in customers purchase decision making to improve cus
4、tomer relationship management and increase sales. In this paper, we present an overview of the impact of social influence in E-commerce decision making to provide guidance to researchers and companies who have an interest in related issues. We identify how data about social influence can be captured
5、 from online customer behaviors and how social influence can be used by Ecommerce websites to aid the user decision making process. We also provide a summary of technology for social network analysis and identify the research challenges of measuring and leveraging the impact of social influence on E
6、-commerce decision making. Categories and Subject Descriptors H.1.2 User/Machine Systems:Human Factors, Human Information Processing； J.1 Administrative Data Processing:Marketing. General Terms Management； Human Factors； Theory. Keywords Social Network； E-Commerce. 1. INTRODUCTION Browsing, searchin
7、g, and buying a product on E-commerce websites is often a time consuming and frustrating task for consumers. Over 80% of Web shoppers have at some point left Ecommercewebsites without finding what they want. Richer E-commerce systems that connect companies to their customers could enhance customers
8、decision making and their bottom line. E-commerce companies are attempting to support part of their potential customers decision making process by introducing personalized Web-based decision support systems such as recommender systems. These recommender systems provide consumers with personalized re
9、commendations based on their purchase history, past ratings profile, or interests. These collaborative filtering based recommender systems have been applied to many E-commerce websites (e.g., movie, music, and restaurant recommendation) and shown good performance in predicting a list of products whi
10、ch a consumer prefers. A typical collaborative filtering algorithm builds a customers neighborhood based on his or her preferences of shared products and weighs the interest of neighbors with similar taste to generate new recommendations . Sinha and Swearingen , however, found that consumers are far
11、 more likely to believe recommendations from people they know and trust, i.e., friends and family-members, rather than from automated recommender systems in E-commerce websites. In reality, a persons decision to buy a product is often strongly influenced by his or her friends, acquaintances and busi
12、ness partners, rather than strangers. Nevertheless, online communities on the Web allow users to express their personal preferences and to share their recommendations by rating others reviews and identifying trusting members. According to new research by Hitwise , social network sites including MySp
13、ace and Facebook are driving an increasing volume of traffic to retail sites(i.e., 6% of retail traffic in 2006), and are thus becoming a starting point for Web users who are interested in E-commerce. This increase in traffic from social network sites to online retailers shows that highly influentia
14、l customers directly affect other consumers decision making. Therefore, E-commerce companies can take advantage of this social influence between consumers to support customer relationship management and increase sales. Approaches incorporating social influence into recommender systems or online mark
15、eting in E-commerce have started gaining momentum. Some researchers have suggested social recommender systems that take into account social interaction in combination with purchase preferences and profiles when generating recommendations. Lam proposed a collaborative recommender system incorporating
16、 social network information,called Social Network in Automated Collaborative-filtering of Knowledge (SNACK). The similarity weight for users ratings was modified according to the network distance (i.e., the length of the shortest path) between two users, and the preference of closednetwork neighbors
17、 was emphasized. Massa et al. built a trust model using trust data from E to predict the trust value that is propagated within a network and used to make recommendations. These social recommender systems have been observed to achieve better prediction rates, and have also solved the cold-start probl
18、em in which a new user does not have enough product preferences for the system to make good predictions. The measure of social influence and trust value in a Web-based social network increasingly appears to be an important key to enhancing the accuracy of recommender systems. Some researchers have f
19、ocused on the consumer networks that are formed through the direct and indirect interactions (e.g., to read and rate reviews) between consumers to maximize the impact of direct marketing through social influence. Models have been proposed to identify a set of highly influential customers to maximize
20、 word-of-mouth effects or to find target customers based on the preferences and influencial impact from previous customers,. Domingos et al. proposed a model to mine a customers network value and optimize the choice of which customers to market to. Kempe et al.solved the optimization problem of sele
21、cting highly influential customers to maximize the spread of influence through a social network. In these efforts, a measure of social influence such as cascading effects and network value is one of the key issues. Hill et al.suggest networkbased marketing using existing customers to identify potent
22、ial customers who are likely to buy, based upon being influenced by previous customers who have bought a service. This was done in the domain of telecommunication services. Although some emerging research has started to incorporate social influence in E-commerce, it has been limited to data sources
23、about social interaction captured from E-commerce interactions only, which is only a subset of the information that is becoming available. In this paper, we present an overview of the impact of social influence in E-commerce decision making to provide guidance to researchers and E-commerce companies.Specifically, we examine various ways to
