Hyperlink Analyses of the World Wide Web: A Review 论文

摘要

We have recently witnessed the growth of hyperlink studies in the field of Internet research. Although investigations have been conducted across many disciplines and topics, their approaches can be largely divided into hyperlink network analysis (HNA) and Webometrics. This article is an extensive review of the two analytical methods, and a reflection on their application. HNA casts hyperlinks between Web sites (or Web pages) as social and communicational ties, applying standard techniques from Social Networks Analysis to this new data source. Webometrics has tended to apply much simpler techniques combined with a more in-depth investigation into the validity of hypotheses about possible interpretations of the results. We conclude that hyperlinks are a highly promising but problematic new source of data that can be mined for previously hidden patterns of information, although much care must be taken in the collection of raw data and in the interpretation of the results. In particular, link creation is an unregulated phenomenon and so it would not be sensible to assume that the meaning of hyperlinks in any given context is evident, without a systematic study of the context of link creation, and of the relationship between link counts, among other measurements. Social Networks Analysis tools and techniques form an excellent resource for hyperlink analysis, but should only be used in conjunction with improved techniques for data collection, validation and interpretation.