Hyperlink Analyses of the World Wide Web: A Review

来源： 作者：unkonwn 时间：2004-12-11 点击：

Our purpose in this paper is to review and compare two approaches to hyperlink analysis, and thereby to contribute to methodological discussions in Internet studies. The approaches and goals of these two areas have some features in common, but also important differences that it will be useful to identify and explain. Both areas have the broad goal of extracting useful information about Web use and employ the general approach of using predominantly quantitative techniques to display or summarize hyperlink-based data. This is in contrast to hyperlink analysis as conducted by statistical physicists and computer scientists, which adopts a much more abstract approach, as will be revealed in the brief review of these areas below.

The two approaches come from different fields. Hyperlink Network Analysis (HNA) derives from Social Network Analysis, whereas Webometrics derives from information science. As a result it would be natural to expect the different backgrounds to be reflected in the kinds of problems tackled, the methods developed and the outcomes sought. Field differences can also mean that there is little interaction between practitioners of each approach, and our review aims to enrich the understanding of these approaches by highlighting what they share and how they differ. 数据挖掘实验室

A second motivation for this review is to introduce the techniques to a wider group of researchers, in the belief that with the increasing importance of the Web for an ever-broader spectrum of human activities, online analysis techniques should be more widely developed and exploited. Hyperlink analysis provides Internet researchers with new analytical methods for the study of networked (or connected) structures on the World Wide Web. This paper both provides a review of hyperlink research that originates in hyperlink network analysis or Webometrics communities and examines practical issues related to link data-collection techniques.

Compared to other Web methods such as a content-based analysis, the relative advantage of hyperlink analysis is that it is able to examine the way in which Web sites form a certain kind of relations with others via hyperlinks. According to Weare and Lin (2000), content-based studies may be missing an opportunity if they fail to consider the hyperlinked environment. Many Web sites with common topics are hyperlinked together, which allows users to access materials or services hosted on other sites. Given this interweaving hyperlinking structure, it may be necessary to recognize individual Web sites as mutually dependent entities, which constitute a Web system. If a content analysis of individual Web sites does not include materials to which the Web site under investigation hyperlinks (such as academic reports that are hosted on other sites), it fails to see the structures in the environment that afford social navigation. Also, hyperlink analysis enables visualization of navigational elements, such as changes in the hyperlink structure of contents of a Web site. It also makes possible the quantification of relational attributes among sites within a community being studied. Using this information in combination with other Web analyses (Howard, 2002; Park, 2002c) can contribute to the understanding of why and how certain types of contents come to appear on Web sites.

Hyperlink Analyses of the World Wide Web: A Review

References