Personalized search has gained great popularity to improve search effectiveness in recent years. The objective of personalized search is to provide users with information tailored to their individual contexts. We propose to personalize Web search based on features extracted from hyperlinks, such as anchor terms or URL tokens. Our methodology personalizes PageRank vectors by weighting links based on the match between hyperlinks and user profiles. In particular, here we describe a profile representation using Internet domain features extracted from URLs. Users specify interest profiles as binary vectors where each feature corresponds to a set of one or more DNS tree nodes. Given a profile vector, a weighted PageRank is computed assigning a weight to each URL based on the match between the URL and the profile. We present promising results from an experiment in which users were allowed to select among nine URL features combining the top two levels of the DNS tree, leading to 29 pre-computed PageRank vectors from a Yahoo crawl. Personalized PageRank performed favorably compared to pure similarity based ranking and traditional PageRank.