The literature on recommendation systems indicates that the choice of the methodology significantly influences the quality of recommendations. The impact of the amount of available data on the performance of recommendation systems has not been systematically investigated. We study different approaches to recommendation systems using the publicly available EachMovie data set. In contrast to previous work on this data set, here a significantly higher subset is used. The effects caused by the number of customers and movies as well as their interaction with different methods are investigated. We compare two commonly used collaborative filtering approaches to several regression models using an experimental full factorial design. According to our findings, the number of customers significantly influences the performance of all approaches under study. For a large number of customers and movies, we show that simple linear regression with model selection can provide significantly better recommendations than collaborative filtering. From a managerial perspective, this gives suggestions about the selection of the model to be used depending on the amount of data available. Furthermore, the impact of an enlargement of the customer database on the quality of recommendations is shown. (author's abstract) / Series: Working Papers SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

Due to a rapid expansion of commercial usage on the Internet, on-line auction is showing its growing potential and economic benefit on business competition.¡@On-line auction is a totally new consuming experience for most people. However, though fast and convenient comparing to traditional shopping, social contacts, such as bargaining, are not provided during the process of on-line shopping.Moreover, stereotyped consuming behavior of " seeing is believing" and concerns on Internet security may also be obstacles for many consumers to shop on-line.Through investigating online auction users' expectation ,trying to find influencing factors for users to adopt on-auction shopping.Factors causing users to avoid on-line shopping are also of great interest.¡@Enterprises doing Internet business may find the results of this study useful in their marketing analysis and planning. A web-based questionnaire was employed.Items were grouped into three sections in the questionnaire: on-audtion shopping environment, on-line experience, and demographic information.¡@A total of 1698 questionnaire was collected. The results indicated that male subjects were higher motivated to shop on-line auction than their female counterparts.Subjects with on- line auction shopping experience were found to have significantly higher motivation to shop on-line auction than those without such experience.Security¡@problem of using credit card for payment via Internet still is the most concerns for many subjects. Provision of detail information regarding product description and procedures for making order, as well as a comprehensive after-sell service may largely encourage web users adopting on-line auction shopping.

Building virtual community can help a web site to increase its traffic flow, make visitor stayed longer, improve customer loyalty and transact motive. In order to start the dynamic circle of increasing return in a virtual community, the virtual community has to create attractive contents. And system operators (SYSOPs) have playing a critical role in accumulating attractive contents. This research is a exploratory research, we try to find out whether there are some potential traits of SYSOPs that will make the contents more attractive. And we hope the research result can provide as a reference for managers of virtual communities in assessing and training their SYSOPs. We discuss SYSOPs¡¦ traits via three dimensions (lifestyle, characteristic, personal statistics) to test whether these traits will influence the attractiveness of contents. Finally, we find out that most of these SYSOPs are male, 16-25 years old, with college education level, whose average monthly income is less than ten thousands, and most of them has experiences in operation another boards. In the lifestyle dimension, after using factor analysis and cluster analysis, we find out the SYSOPs¡¦ lifestyle can be divided into five types, there are group of X generation, group of independent, group of elegant, group of passive and group of aggressive. But there wasn¡¦t any significant difference between them. In the dimension of characteristic, among three variables (social technique, affiliation, and aggressiveness) there wasn¡¦t any significant difference between the group of high performance SYSOPs and the group of low performance SYSOPs. Finally in the test of personal statistics, we found that experience of operating other boards will provide a significant influence in SYSOPs¡¦ performance. And we use this variable to form a discriminate function, but after test of wilks¡¦ £N, the discriminate ability of this discriminate function wasn¡¦t reach the significant level. So it can not provide an effective forecast for the performance of SYSOPs in virtual communities.

Thesis (Ph. D.)--University of Texas at Austin, 1999. / Vita. Includes bibliographical references (leaves 292-305). Available also in a digital version from Dissertation Abstracts.

Thesis (Ph. D.)--University of Texas at Austin, 1999. / Vita. Includes bibliographical references (leaves 115-121). Available also in a digital version from Dissertation Abstracts.

Thesis (M.S.)--University of Florida, 2001. / Title from first page of PDF file. Document formatted into pages; contains x, 57 p.; also contains graphics. Vita. Includes bibliographical references (p. 54-56).

Thesis (M.S.)--West Virginia University, 2002. / Title from document title page. Document formatted into pages; contains v, 42 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 27-28).

Zugl.: Hamburg, Univ., Diss., 2007

Thesis (Ph. D.)--University of Texas at Austin, 1999. / Includes bibliographical references (leaves 115-121). Available also in a digital version from Dissertation Abstracts.

Thesis (Ph. D.)--University of Texas at Austin, 2001. / Vita. Includes bibliographical references. Available also from UMI/Dissertation Abstracts International.