This web mining paper aims at analyzing whether the number of links pointing to a web page is biased by the structure of web sites.
By web‐design mining methods, two collections of web pages are extracted and the in‐links counts are determined by querying web search engines.
The paper finds that the structure bias and pages on a higher hierarchical level are likely to receive more links than other pages.
The links are set by web page authors whose behaviour is not yet fully understood and which requires further research.
The paper shows that the structure bias of in‐links should be considered by link analysis measures used in search engines. Potential consequences are presented.
The number of links toward a web page are considered to be an indicator for the quality of that page. This measure is biased and cannot be solely trusted as a knowledge source for the quality of a page.
